Forensic Dna Evidence For Criminal Justice Professionals: Introduction To

Introduction to
FORENSIC DNA
EVIDENCE FOR
CRIMINAL JUSTICE
PROFESSIONALS
Introduction to
FORENSIC DNA
EVIDENCE FOR
CRIMINAL JUSTICE
PROFESSIONALS
Jane Moira Taupin
Boca Raton London New York
CRC Press is an imprint of the

Taylor & Francis Group, an informa business
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
2014 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works

Version Date: 20130503
International Standard Book Number-13: 978-1-4398-9910-6 (eBook - PDF)
This book contains information obtained from authentic and highly regarded sources. Reason-
able efforts have been made to publish reliable data and information, but the author and publisher
cannot assume responsibility for the validity of all materials or the consequences of their use. The
authors and publishers have attempted to trace the copyright holders of all material reproduced in
this publication and apologize to copyright holders if permission to publish in this form has not
been obtained. If any copyright material has not been acknowledged please write and let us know so
we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information
storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.
copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc.
(CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organiza-
tion that provides licenses and registration for a variety of users. For organizations that have been
granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and
are used only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Contents
Preface..................................................................................................................xi
Acknowledgments............................................................................................xv
About the author............................................................................................ xvii
Chapter 1 History of forensic DNA profiling in criminal

investigations................................................................................ 1
1.1 Discovery of structure and importance of DNA molecule: A
Nobel prize................................................................................................ 1
1.1.1 Mendelian law of inheritance................................................... 1
1.1.2 Structure of DNA........................................................................ 2
1.1.3 Human genome project............................................................. 3
1.2 DNA and concept of individuality........................................................ 4
1.3 Alec Jeffreys and the worlds first murder case solved by DNA....... 5
1.4 Early criminal court challenges to DNA technology.......................... 7
1.5 Changing the face of forensic science: The value of biological
evidence..................................................................................................... 9
References.......................................................................................................... 11
Chapter 2 Strengths and limitations of DNA profiling evidence...... 13

2.1 Introduction: Power and caution.......................................................... 13
2.2 Discrimination power of DNA profiling............................................. 15
2.3 Genetic basis for DNA profiling........................................................... 16
2.4 Stability of DNA profiling..................................................................... 18
2.5 Persuasive statistics................................................................................ 19
2.6 Relatives................................................................................................... 21
2.7 DNA databases....................................................................................... 22
2.8 DNA intelligence-led policing.............................................................. 23
2.9 Mass disasters......................................................................................... 24
2.10 DNA evidence in context....................................................................... 24
2.11 Time of deposition: Transfer and persistence of DNA...................... 25
v
vi Contents
2.12 Relevant evidence?................................................................................. 27

2.13 Relevant exhibits?................................................................................... 29
2.14 CSI effect and the notion of infallible forensic evidence.............. 30
2.15 Relationships of lawyers and scientists............................................... 30
References.......................................................................................................... 30
Chapter 3 DNA profiling basics................................................................ 33

3.1 What is DNA?.......................................................................................... 33
3.2 Biological materials allowing DNA profiling..................................... 33
3.2.1 Searching for DNA on exhibits.............................................. 35
3.2.2 Blood........................................................................................... 37
3.2.3 Semen and spermatozoa.......................................................... 38
3.2.4 Saliva.......................................................................................... 39
3.2.5 Hair roots................................................................................... 40
3.2.6 Dandruff and skin.................................................................... 40
3.2.7 Nasal secretions........................................................................ 40
3.2.8 Vaginal secretions..................................................................... 40
3.2.9 Sweat........................................................................................... 41
3.2.10 Wearer DNA.............................................................................. 41
3.2.11 Touch DNA................................................................................ 42
3.2.12 Urine and feces......................................................................... 43
3.2.13 Emerging techniques............................................................... 43
3.3 Reference samples.................................................................................. 44
3.3.1 Buccal scrapes........................................................................... 44
3.3.2 Blood........................................................................................... 44
3.3.3 Plucked hair samples............................................................... 44
3.3.4 Personal belongings................................................................. 45
3.4 Current profiling technique: Short tandem repeats (STRs).............. 45
3.5 Reading tables of alleles........................................................................ 47
3.6 Obtaining DNA profiles........................................................................ 51
3.6.1 Controls...................................................................................... 51
3.6.2 Extraction................................................................................... 52
3.6.3 Quantification........................................................................... 52
3.6.4 Amplification............................................................................. 53
3.6.5 Separation and detection......................................................... 54
3.6.6 Reading electropherograms.................................................... 54
3.6.7 Artifacts and other technical issues....................................... 55
3.7 Time required to obtain DNA profiles................................................ 59
3.8 Designating peaks.................................................................................. 60
3.9 Case documentation and review.......................................................... 61
References.......................................................................................................... 63
Contents vii
Chapter 4 Evidential value and statistics................................................. 65

4.1 Introduction............................................................................................. 65
4.2 Interpreting DNA profiles..................................................................... 66
4.3 Statistical approaches and obtaining final statistics......................... 66
4.3.1 Random match probability and likelihood ratio................. 66
4.3.2 Calculating frequencies........................................................... 68
4.3.3 Comparison of probability of exclusion and LR methods.... 69
4.3.4 Identity and rarity.................................................................... 71
4.4 Legal fallacies.......................................................................................... 71
4.5 Understanding reports: Common phrases and their meanings...... 73
4.5.1 Inclusion and exclusion........................................................... 73
4.5.2 Declared contributor................................................................ 75
4.5.3 Verbal descriptors..................................................................... 76
4.6 Sampling correction and uncertainty.................................................. 76
4.7 Relevant population and impact on statistical value........................ 77
4.8 Relatives................................................................................................... 77
References.......................................................................................................... 78
Chapter 5 Partial profiles, low levels, and mixtures............................. 81

5.1 Partial profiles......................................................................................... 81
5.2 Low level and suboptimal profiles....................................................... 82
5.2.1 Definitions................................................................................. 82
5.2.2 Stochastic effects....................................................................... 85
5.2.3 An interesting experiment...................................................... 88
5.2.4 Enhancement techniques........................................................ 90
5.2.5 Improving reliability of results.............................................. 90
5.2.5.1 Biological (consensus) model................................. 91
5.2.5.2 Statistical (probabilistic) model.............................. 91
5.2.6 Contamination.......................................................................... 92
5.3 DNA mixtures from two or more people........................................... 93
5.4 Mixture interpretation steps................................................................. 98
5.5 Low template mixtures.......................................................................... 99
5.6 Complex mixtures................................................................................ 100
References........................................................................................................ 101
Chapter 6 Y-STR profiling........................................................................ 105

6.1 Introduction........................................................................................... 105
6.2 Benefits................................................................................................... 106
6.3 Theory.................................................................................................... 109
6.4 Statistics..................................................................................................110
6.4.1 Frequency estimates of Y -STR haplotypes...........................110
6.4.2 Meaning of Y-STR match........................................................111
viii Contents
6.5 Number of male contributors to Y-STR profile.................................113

6.6 Determining mixture ratios.................................................................113
6.7 Combining statistics from autosomal and Y-STR profiling............114
References.........................................................................................................114
Chapter 7 Other DNA techniques including mitochondrial DNA.... 117

7.1 Introduction............................................................................................117
7.2 DNA analysis of bone...........................................................................117
7.3 Mitochondrial DNA basics...................................................................119
7.4 Statistics in mitochondrial DNA analysis......................................... 122
7.5 Contamination...................................................................................... 124
7.6 Mixture mitochondrial DNA profiles................................................ 125
7.7 Familial DNA searching...................................................................... 126
7.8 Domestic animal hair........................................................................... 128
7.9 Other techniques.................................................................................. 129
References........................................................................................................ 129
Chapter 8 Concerns and controversies................................................... 131

8.1 Introduction........................................................................................... 131
8.2 Quality issues........................................................................................ 132
8.3 Relevant sample testing....................................................................... 133
8.4 Contamination...................................................................................... 134
8.5 Interpretation issues............................................................................. 138
8.6 Error rates.............................................................................................. 138
8.7 Overreliance on DNA technology...................................................... 140
8.8 Interpretation of DNA profiles: Objectivity and subjectivity.........141
8.9 Retesting of samples............................................................................. 143
8.10 Adversarial system............................................................................... 144
8.11 Misconception about exact science.................................................... 144
8.12 Obligations............................................................................................. 144
References........................................................................................................ 145
Chapter 9 DNA pointers for criminal justice professionals.............. 147

9.1 Introduction........................................................................................... 147
9.2 Advantages of DNA profiling............................................................. 148
9.3 Querying DNA evidence: Advice for the prosecution and the
defense................................................................................................... 148
9.4 Warning signs....................................................................................... 149
9.5 Was all evidence tested?...................................................................... 150
9.6 Pretrial review....................................................................................... 151
Contents ix
9.7 Suggested cross-examination questions........................................... 151

9.7.1 General..................................................................................... 151
9.7.2 Single source DNA profiles associated with blood,
semen, or saliva....................................................................... 152
9.7.3 Difficult DNA profiles (partial, low level, mixture,
unspecified origin)................................................................. 152
9.7.4 Expert witness......................................................................... 153
9.8 Discovery requests............................................................................... 153
Appendix A: Glossary of terms used in reports and testimony.............. 155

Appendix B: Selected DNA issues and case examples...............................161
Appendix C: Steps in review of evidence.................................................... 163
Index................................................................................................................. 165
Preface
Purpose of this book

This book aims to provide trial lawyers and criminal justice profession-
als with sufficient tools to understand and probe the evidential value
of forensic DNA evidence in their criminal cases. This text is designed
for criminal lawyers and criminal justice professionals and written for
nonscientific readers.
The criminal justice professional will gain specific knowledge of the
strengths and limitations of DNA evidence in criminal cases. The pros-
ecution lawyer will improve his or her understanding of DNA evidence
when such evidence should be emphasized, when to discuss the evidence
with forensic experts, and when to proceed with trial despite a lack of
DNA evidence. The defense lawyer may be better equipped to chal-
lenge DNA evidence and perhaps employ an independent expert, under-
stand when to focus on other aspects of the prosecutions case, and know
when to secure the advantage of an early guilty plea.
The extreme probabilities quoted in many criminal cases with DNA
evidence may make it appear that such evidence allows no margin for
error. Many lawyers do not even know what one in one trillion means,
yet this number is cited in many forensic reports. However, recent cases
worldwide have shown that DNA evidence is not infallible. There is a
danger in relying on DNA statistical probabilities in the determination of
guilt. DNA evidence is just one piece of circumstantial evidence and does
not prove an accused was the offender. Also, DNA profiling is performed
and interpreted by humans and thus, like all scientific testing, is subject
to quality control and other errors.
DNA profiling for use in forensic cases was a significant scientific
achievement in the 1980s and has been considered the most innovative
technique in forensic science since fingerprinting. Yet how can a lawyer
with limited scientific knowledge grasp the technology and exploit it
or challenge it? This text aims to provide criminal lawyers with the
xi
xii Preface
knowledge to do just that and make DNA profiling less intimidating.

DNA profiling should not be a Pandoras box that both prosecution and
defense attorneys are afraid to open for fear of what may be unleashed.
All sides of the adversarial system should be confident and aware of the
strengths and limitations of DNA evidence in order for the criminal jus-
tice system to operate effectively. This book aims to assist in this process
so that a criminal lawyer can confidently understand and convey this
innovative technology in a courtroom.
DNA offers a degree of certainty often missing in a criminal trial.
It helps focus police investigations and solve decades-old cold cases.
Alternatively, the investigating police may be happy with a DNA result
and become complacent, and thus fail to explore or reject other avenues
of evidence. Counsel for both parties should be aware of biased or sloppy
investigations. Similarly, forensic scientists, judges, and jury members
should be aware of the limitations and strengths of DNA evidence. This
book aims to provide readers with the tools to recognize both aspects
of testing. The CSI effect may impact the judge, jury, and even legal
practitioners so that the DNA evidence is accorded more weight than it
warrants. Again, this text hopes to provide a balanced perspective on the
weight of DNA evidence in criminal trials.
The understanding of DNA concepts by the jury is also an issue.
Ultimately, it is up to legal counsel to convey the concepts of DNA profil-
ing to a jury in a manner that is readily understood. This text is intended
to provide the tools for criminal lawyers to use DNA profiling effectively.
Scope and limitations

This book is designed for lawyers and other criminal justice profession-
als (such as legal researchers) who have limited scientific knowledge.
However, forensic scientists and those interested in the application of
DNA to criminal proceedings should also find it valuable. It should be
thought-
provoking reading for crime authors, journalists, and legal
commentators.
Most of the background scientific information needed to understand
the basis of DNA profiling is provided in this text. Although some con-
cepts may appear too complex or scientific on initial perusal, readers are
encouraged to grasp the concepts with which they are comfortable and
engage in further reading and/or discussions among colleagues.
Literature references to scientific papers are provided but it is not nec-
essary for readers to peruse any or all of them. However, readers should
be aware of the references because of their potential to be mentioned in
court. A forensic scientist providing DNA evidence at trial should be famil-
iar with some, if not most, of the literature quoted, especially the most
Preface xiii
famous papers cited in this book. The references are intended to assist
trial lawyers in their examinations of expert witnesses. Understanding
and quoting the literature will enable them to assess an expert witnesss
knowledge of the matters discussed therein.
The text covers the most common DNA methods used in criminal
trials todaynuclear DNA short tandem repeat (STR) techniques, mito-
chondrial DNA, and Y-STR profiling. The statistics discussed are the
probabilities obtained by comparisons of reference samples and crime
scene samples. Paternity statistics will not be explored.
The book is applicable to the adversarial (trial and jury) legal sys-
tems encountered in countries such as England, the United States, and
Australia. Many of the principles can be applied in other countries that do
not utilize juries but use the inquisitorial approach (courts in Europe con-
sist of panels of judges and a single judge presides over court proceedings
in Middle Eastern countries).
Since this text does not discuss the forensic principles in depth or
assume scientific knowledge on the part of readers, it contains numerous
references to specific forensic science and statistical texts that can provide
more complete discussions on particular topics.
There are also references throughout the book to case studies from
around the world that illustrate one or more particular applications of
DNA profiling evidence to criminal proceedings and whether that evi-
dence has been applied effectively. The intent is to provide legal profes-
sionals with pointers that may be applicable to their own cases. The final
chapter includes a handy list of questions for a criminal justice profes-
sional to consider when trying a case involving the use of DNA evidence.
Acknowledgments
I thank Becky Masterman, senior editor at Taylor & Francis Group for her
support and enthusiasm for this topic and, as always, her endless patience.
I thank my former colleagues at LGC Forensics England for their advice
on DNA profiling when I moved there from Australia. Special thanks go
to Pauline Stevens and Craig Davies, LGCs DNA gurus. I also thank
Craig for his continuing advice in the advancing field of DNA profil-
ing evidence.
Thanks also go to Dr. Roland van Oorschot at the Victoria Police
Forensic Services Department in Melbourne for his support while I was at
this laboratory, especially for his help in publishing case studies. His own
continuing flood of publications in the DNA literature shows his enthusi-
asm for his specialty.
I appreciate the suggestions of Victoria barrister Peter Chadwick, S.C.
indicating what a barrister would like to see in a DNA book. I also thank
the many barristers and lawyers with whom I worked, who devoted con-
siderable time and effort to understanding a topic that is often alien to legal
personnel. Special thanks go to New South Wales Public Defenders Ian
Nash and Richard Wilson, and Victoria barristers Alan Hands, Benjamin
Lindner, Moya OBrien, Carmen Randazzo, and Samantha Poulter.
This book is for all the criminal justice professionals who are con-
fronted with DNA reports and may have little time to understand them or
prepare cases involving DNA evidence. I hope you find it useful.
xv
About the author
Jane Moira Taupin obtained a BS (Honours) from the University of
Melbourne in Australia. After graduating, she accepted research posi-
tions at the University of Melbourne, first in antibody production at
the Howard Florey Institute and then in cancer research at the Austin
Hospital. She joined the Australian Federal Police as a Constable and
advanced to Stage 1 Detective, working in diverse areas including drug
surveillance and government fraud. She was transferred temporarily to
the only atomic energy facility in the country (Lucas Heights) and used
neutron activation analysis on a number of criminal cases.
She left to join the Victoria Police Forensic Services Centre as a forensic
scientist where she pursued a wide variety of major crime cases involving
biological evidence. Taupin investigated crime scenes for blood pattern
analysis and conducted searches for biological fluids. She has presented
biological expert evidence in courts of law since 1987 and has presented
DNA profiling evidence in court since 1999.
Taupin earned a postgraduate diploma along with an MA, both in
criminology from the University of Melbourne. Her masters thesis in 1994
on the impact of DNA profiling was one of the first in the field.
She then moved to Forensic Alliance in England where she performed
similar work in the companys Oxford and Manchester laboratories.
When LGC Forensics took over the company, she became a lead scien-
tist. In December 2009, she returned to MRS Limited in Melbourne as an
international forensic auditor. She also lectured in Qatar and Bahrain on a
variety of subjects including DNA analysis. She is currently an indepen-
dent forensic consultant and trainer.
Taupin has published many articles in p eer-reviewed journals dis-
cussing trace evidence, clothing damage, and blood pattern analysis. She
also co-authored a text on the forensic examination of clothing.
She won a Young Investigators Award from the International
Association of Forensic Sciences and attended its meeting in Tokyo in
1996 in recognition of her work on clothing damage analysis. The follow-
ing year she won a Michael Duffy travel fellowship from the Australian
xvii
xviii About the author
government to attend the American Academy of Forensic Sciences meet-

ing in New York and visit international laboratories including the FBI in
the United States, the Forensic Science Service in England, and the BKA
in Germany. She participated on the inaugural committee of the Scientific
Working Group on Hair (SWGHAIR) under the auspices of the FBI in
Washington for 6 years. In 2009, she received a Good Citizen Award from
the Greater Manchester Police in England for her work in helping to solve a
horrific case of rape of an elderly woman through DNA profiling evidence.
chapter one
History of forensic DNA profiling

in criminal investigations
Much of the history of deoxyribonucleic acid (DNA) is truly fascinating
and as compelling as any b est-selling crime novel. Scientists, particularly
those in the biology discipline, are familiar with the breakthroughs that
led to the use of DNA in criminal work. They use these stories as build-
ing blocks to assess new information; the criminal justice professional
may benefit from these stories as well. This chapter will briefly introduce
the legal professional to some important discoveries and famous scien-
tists who unlocked the secret of how life may work and how the key was
applied to forensic casework. Scientific terminology used will be more
fully explained in later chapters.
1.1Discovery of structure and importance

of DNA molecule: A Nobel prize
1.1.1Mendelian law of inheritance
The history of DNA can be thought of as beginning with an Austrian
(now Czech Republic) Augustinian monk, Gregor Mendel, who is called
the founder of genetics. In 1865, he completed a series of experiments with
peas and showed that certain traits, such as shape and color, were inher-
ited in different packagesnow called genes. When crossing white flower
and purple flower plants, Mendel found that the result was a purple
flower, not a blend of the colors. He conceived the idea of heredity units
that were either recessive or dominant (Mendel, 1865). These units (genes)
normally occur in pairs in body cells but segregate during the formation
of sex cells. A dominant gene will hide a recessive gene.
Mendel stated that each person inherits two traits from each parent. If
the traits are the same, they are homozygous; if they are different, they are
heterozygous. This becomes an important factor for forensic scientists and
lawyers when interpreting DNA profiles. The alternative forms of each
trait are called alleles.
The two principles Mendel described were the Law of Segregation,
by which each parent passes a randomly selected allele to its offspring
1
2 Introduction to forensic DNA evidence for criminal justice professionals
during fertilization, and the Law of Independent Assortment, by which

separate genes for separate traits are passed independently from parent to
child. Of the 46 chromosomes in a human cell, half are from the mothers
egg and the other half from the fathers sperm. The sex chromosomes X
and Y are incorporated into every DNA profile.
1.1.2Structure of DNA
Until a scientific paper was published in 1944 (Avery et al.), biologists
thought that genesthe units of inheritancewere made of proteins.
Oswald Avery, an American scientist, managed to transfer the ability to
cause disease from one strain of bacteria to another, showing the con-
nection between nucleic acids and genes. This paper has been described
by Nature, a leading scientific journal, as the defining moment in nucleic
acid research.
The following decade saw the amazing discoveries of the structure of
DNA and how it is copied from one generation to the next. Linus Pauling
in California postulated a triple helical structure for DNA in 1953. So had
James Watson and Francis Crick working at Cambridge in England, but
they were all wrong. It was Rosalind Franklins x-ray diffraction photo-
graph that was (unbenownst to her) shown to James Watson that revealed
the true structure of DNA to Watson and Crick.
Nature considers the year 1953 an annus mirabilis (year of wonders) for
science. The three-dimensional structure of DNA was first described by
Watson and Crick in April 1953 in a Nature article. This was the first expla-
nation of how genetic information is encoded and transferred from one
generation to the next. This classic paper first describes the double helical
structure of DNA. Nature later stated that the authors noted with some
understatement that the structure suggests a possible copying mecha-
nism for the genetic material. Another paper in the same issue of Nature
analyzes the x -ray crystallography evidence and suggests that a double
helical structure exists in biological systems (Wilkins et al., 1953).
Following on from this, Rosalind Franklin and Ray Gosling, her stu-
dent, provided further evidence of the helical natures of nucleic acids and
concluded that their phosphate backbones lie on the outsides of the struc-
tures (Franklin and Gosling, 1953). In the next months issue of Nature,
Watson and Crick followed up with largely accurate speculation on how
the base pairing in the double helix allows replication of DNA (Watson
and Crick, 1953).
Watson and Crick and Maurice Wilkins were awarded the Nobel
Prize in Physiology and Medicine in 1962 for their discovery of the double
helix structure of the DNA molecule (Figure1.1). A b est-selling memoir by
James Watson (1968) gives his account of the efforts to beat Linus Pauling
Chapter one: History of forensic DNA profiling in criminal investigations 3
Base pairs
Adenine Thymine
Guanine Cytosine
Sugar phosphate
backbone
U.S. National Library of Medicine
Figure 1.1 Structure of DNA. The two strands of the double helix are connected
by base pairs. The bases are adenine and thymine, and guanine and cytosine.
(Source: U.S. National Library of Medicine.)
to solving the structure of life. Watson was only 24 when he made the
discovery and the book portrays scientists as eminently human, with
petty rivalries and driving ambitions. Many people today believe that
Rosalind Franklin also should have been awarded the Nobel Prize but she
was never even nominated, and died of cancer at age 37 in 1958. Watson,
in an epilogue to his memoir, acknowledges his incomplete and unjust
depiction of Franklin. He later helped establish the Human Genome
Project (see next section).
1.1.3Human genome project

One of the misconceptions that the public may carry is that DNA profil-
ing examines all areas of the DNA molecule. This is currently impossible
to do for every criminal case. In fact, the entire set of genes in DNA had
never been identified until long after the Human Genome Project com-
menced in 1980. The projects goal was to identify the 20,000 or more
genes and determine the sequences of the 3 billion base pairs that make
up human DNA. A genome consists of the entire DNA in an organism
including its genes.
After 13 years of laborious and meticulous work, the Human Genome

Project was completed in 1993. It was coordinated by the U.S. Department
of Energy and National Institutes of Health and the Wellcome Trust in
the U.K. Additional contributions came from Australia, Japan, France,
Germany, China, and other nations.
It is possible that when DNA sequencing techniques progress further,
direct comparisons of very large DNA segments may be done. It is even
conceivable that comparisons of whole genomes may be feasible and thus
allow precise individual identification in the future. At the moment, how-
ever, we are reliant on the observations of a designated number of areas
on the DNA molecule and statistical analysis in the comparison between
the individual and an evidence sample. Furthermore, each individuals
DNA is approximately 99.9% the same as anothers, so that we are con-
cerned with only very small differences in their sequences.
1.2DNA and concept of individuality

Today most people understand that DNA makes every individual
uniqueexcept for identical (monozygotic) twins who emerge from the
same egg. The use of the term DNA profiling without a prefix generally
implies autosomal STR profiling. This is the analysis of the DNA in the
nucleus of the cell from chromosome to chromosome (autosomal relates
to chromosomes) and is the most common DNA profiling method used
in criminal cases. The term STR is discussed in Chapter 2. Y-STR pro-
filing is a specific kind of nuclear DNA profiling which analyzes the Y
chromosome only, is inherited paternally and is much less discriminatory
(discussed in Chapter 6). Mitochondrial DNA (mtDNA) profiling analyzes
the DNA in mitochondria located outside the nucleus of the cell, is also
less discriminatory and is inherited maternally (discussed in Chapter 7).
DNA is a complex chemical considered to be a genetic blueprint that
determines our chemical and physical characteristics. The genes carry
information for making all the proteins required by an organism. These
proteins determine how an organism looks, how well it fights infection,
and sometimes how it behaves. Other sequences of DNA have structural
purposes or are involved in regulating the use of this genetic information.
As described above, half of human DNA is inherited from the mother
(from the egg that is fertilized) and half from the father (from spermato-
zoa). Forensic DNA profiling examines locations along the DNA molecule
that are highly variable from one individual to another. These regions are
considered junk DNA segments that appear to have no known function in
the human body.
DNA is a nucleic acid consisting of two long polymers called nucleo
tides, with backbones made of sugars and phosphate groups. The two
polymer strands run in opposite directions and form the shape of the
double helix as first described by Watson and Crick. Attached to each sugar
molecule is one of four similar chemicals called bases. These bases are
called adenine, thymine, guanine, and cytosine and are repeated billions
of times throughout a genome. The particular order of the bases is impor-
tant and is responsible for lifes diversity. These bases also become impor-
tant in the interpretations of DNA profiles in criminal cases.
DNA within a cell is organized into long structures called chromo-
somes. These chromosomes are duplicated before cells divide in a pro-
cess called DNA replication. It is naturally accepted in recently published
papers that the main precondition for the transfer of genetic information
from one cell to its daughter cells and from one generation to the next is
the high stability of DNA (Vennemann and Koppelkamm, 2010).
1.3Alec Jeffreys and the worlds first

murder case solved by DNA
When this author commenced work as a forensic scientist in 1986, the
concept of DNA profiling in criminal cases was unknown. Conventional
genetic markers such as ABO blood grouping and enzyme typing were
used on physiological fluid stains such as blood and semen. These mark-
ers used the theories of population genetics that DNA analysis also uses;
however these markers have very low discrimination power. For example,
the blood group O is common to nearly half the population. Often due
to the ages and sizes of stains on clothing or weapons, limited or no infor-
mation is obtained due to the instability of the enzymes and blood group-
ings and the susceptibility to bacterial attack. It was frustrating to forensic
scientists to find noticeable quantities of blood and semen on exhibits
from crime scenes and obtain no evidential value from them. These early
serology tests have now been superseded in most forensic laboratories.
DNA profiling was introduced into the forensic arena by notable pub-
lications from English scientists, again in Nature (Jeffreys et al., 1985; Gill
et al., 1985). Professor (now Sir) Alec Jeffreys of Leicester University discov-
ered in 1984 that VNTRs (variable numbers of tandem repeats) were pres-
ent in all human DNA but varied in length for each individual. He called
his method DNA fingerprinting because it typed many minisatellites
simultaneously so that the final result resembled a distinct barcode.
This technique, using restriction fragment length polymorphism
(RFLP), was used first in a disputed immigration case to confirm the iden-
tity of a British boy whose family was originally from Ghana (Jeffreys
et al., 1986). The case was resolved when DNA results showed the boy
was closely related to other members of the family. The first murder case
solved by DNA profiling (Case 1) was prosecuted in the English midlands

in the mid 1980s (Gill and Werrett, 1987; Wambaugh, 1989).
Case 1
Lynda Mann, a 15-year-old girl, was found raped and mur-
dered in 1983. Three years later, 15-year-old Dawn Ashworth
was raped and murdered nearby. A kitchen porter confessed
to the second murder but not the first; police, however, were
convinced both girls had been murdered by the same offender.
The semen on both bodies fell into the type A blood group and
the enzyme profile matched 10% of the adult male population.
The police asked Professor Alec Jeffreys to analyze the samples
using his new technique of DNA fingerprinting.
The semen on both bodies was indeed from the same man
but DNA profiling excluded the kitchen porter. This led the
police to conduct a world firsta DNA screen of 3,000 local
men. Colin Pitchfork persuaded a work colleague to donate
a sample for him, but police discovered the ruse and subse-
quently found that the DNA profile of the semen on the bodies
matched Pitchforks DNA profile. Pitchfork was convicted and
sentenced for the two murders in 1988.
Professor Jeffreys was only 34 years old at the time, a young genius like
James Watson. His work on the murder case was confirmed by Dr.Peter
Gill and Dr.Dave Werrett of the Forensic Science Service (FSS) in England,
who jointly published the first paper on applying DNA profiling to foren-
sic science (Gill et al., 1985). The first suspect in the Pitchfork case was
thus the first to be exonerated by DNA profiling. The high discrimination
powerthe power to excludearguably produced the greatest impact of
DNA profiling in the criminal justice field.
The Pitchfork case led to the rapid introduction of DNA analysis into
casework in England and Wales (Werrett et al., 1989). After the polymerase
chain reaction (PCR) technology was developed in the late 1980s, DNA
typing methods began incorporating PCR and RFLP technology was
phased out. PCR is essentially a molecular photocopier that can amplify
very small samples into quantities that can be detected and analyzed. This
method is a boon for forensic science as it enables the analysis of minute
quantities of blood and semen, as well as degraded samples such as those
commonly encountered at crime scenes. Today, microsatellites are used
instead of minisatellites.
Professor Jeffreys first exploited DNA analysis to confirm the iden-

tity of Josef Mengele (Jeffreys et al., 1992), the notorious Nazi concentra-
tion camp physician who performed horrific experiments on Auschwitz
inmates. The DNA from an exhumed femur bone of the presumed skel-
eton of Mengele (buried in Brazil as Wolfgang Gerhard) was compared to
the DNA from his widow and son.
The DNA fingerprinting term has been dropped in favor of DNA pro-
filing to eliminate the poor analogy with fingerprint technology, which
has no genetic statistical basis. Chapter 3 discusses the current profil-
ing techniques.
1.4Early criminal court challenges

to DNA technology
One of the first court cases in the United States using DNA profiling was
conducted in November 1987 in Orange County, Florida. Tommy Lee
Andrews was convicted of rape based on testing that matched DNA from
his blood with DNA from semen taken from the victim (Aronson, 2007).
The prosecution merely had to prove the reliability of the evidence, not
general scientific acceptance. Aronsons book describes how quality
assurance and control were lacking in the early stages of DNA profiling
and how subsequent defense challenges improved quality at all stages of
the process including collection and sampling.
The West Virginia Supreme Court was the first state high court to
rule on the admissibility of DNA evidence in the case of State v. Woodall in
1987. The court accepted DNA testing by the defendant but inconclusive
results failed to exculpate Woodall. The court upheld the conviction for
rape, kidnapping, and robbery of two women but Woodall continued to
pursue further testing. DNA analysis using PCR determined that he was
innocent and he was released from prison in 1992. The real perpetrator
was eventually found (Innocence Project).
The first case that seriously challenged the admissibility of DNA evi-
dence was People v. Castro heard in the New York Supreme Court. Castro
was accused of murdering his neighbor and her 2-year-old daughter
(National Research Council, 1992). Blood found on Castros watch con-
tained DNA that matched the DNA of the dead woman but Castro claimed
the blood was his own. During a pretrial hearing, the court held that DNA
was generally accepted in the scientific community but that the technique
as applied in the Castro case was so flawed that the evidence of a match
was not admissible, although evidence indicating that the DNA was not
Castros was admissible. The court concluded that pretrial hearings are
required to determine whether a testing laboratorys methods meet sci-

entific standards and are reliable. The defendant subsequently pleaded
guilty. As a result of this case, the scientific and legal communities raised
concerns about DNA profiling in the United States.
Pretrial hearings may also be held in England to decide the admissi-
bility of DNA evidence in question. In the case of R. v. Doheny and Adams
(1997), the court noted that the risk of laboratory error, the method of DNA
analysis used, and the basis of statistical calculations should be examined.
The O.J. Simpson trial (Case 2) focused the attention of the American
public on forensic evidence and DNA in particular.
Case 2
Former American football star and actor O.J. Simpson was tried
on two counts of murder following the June 1994 deaths of his
ex-wife Nicole Simpson and her friend Ronald Goldman. The
trial was held in Los Angeles from January to October 1995.
It has been described as the most publicized criminal trial in
American history and educated a generation of Americans on
the potential of DNA evidence. The trial included 133 days of
televised testimony (available on the CNN website; People v.
Simpson, Cal. Sup. Ct., 1995).
The defense team persuaded the jury that there was reason-
able doubt about the DNA evidence and cited mishandling of
evidence by the Los Angeles police and sloppy internal labora-
tory procedures that contaminated the evidence. In 1997, a civil
suit was brought against O.J. Simpson by the family of Ronald
Goldman (Goldman v. Simpson, Cal. Sup. Ct.). The outcome was
a $35 million wrongful death judgment against Simpson.
Orchid Cellmark was the independent laboratory that
tested more than 100 pieces of bloodstained evidence (Orchid
Cellmark). In a different matter, O.J. Simpson was arrested in
2007 in Las Vegas and charged with armed robbery and kid-
napping. He was found guilty and sentenced to a minimum of
9 years in jail without parole.
Issues such as the novelty, reliability, and validity of routine DNA pro-
filing in criminal trials in the twenty-first century are rarely debated.
However, on occasion, other issues may be explored by defense counsel.
Over a decade ago, this author was required to serve as an expert wit-
ness in a contested trial involving DNA evidence in Melbourne, Australia
(Case 3; Taupin, 2001).
Case 3
DNA was found on a plastic bag, business card, and a card-
board box containing ecstasy tablets that linked the accused to
drug trafficking from the state of Queensland to Melbourne.
The defense objected to the presentation of the evidence as
hearsay, as the author of the report had not done all of the test-
ing, and thus requested a voir dire examination on the admis-
sibility of the evidence. In fact, the routine analysis of DNA
evidence can involve six or more scientists and it is rare for the
reporting scientist to be involved in all stages of the process.
After cross-examination by the defense on many aspects
of DNA covering details of extraction and analysis, the judge
admitted the evidence and the case proceeded to trial with the
proviso that the first and second typers (who initially exam-
ined the DNA profile electropherogram) were to give evidence.
The jury ultimately delivered a verdict of guilty.
Many more case studies illustrating other aspects of DNA evidence in crim-
inal trials will be discussed throughout subsequent chapters in this text.
1.5Changing the face of forensic science:

The value of biological evidence
These highly publicized cases using DNA profiling led to an explosion in
the use of DNA technology in criminal cases. This author has personally
seen the value of analyzing biological materials in violent criminal cases
increase exponentially since the introduction of such testing in cases han-
dled by government and private forensic laboratories. The authors MA
thesis (Taupin, 1994) written early in the history of the use of DNA testing
in criminal trials, however, showed that DNA is relevant in only a small
proportion of criminal proceedings because of a lack of DNA evidence or
because a matter is not contested (consent in sexual offenses, for example).
Much of the impact is behind the scenes, when DNA profiling is used to
exclude suspects or aid plea bargaining.
DNA profiling technology is constantly evolving and improving as
science continues to advance. The PCR innovation, for example, drasti-
cally changed the value of small and often degraded crime samples.
Materials that cannot be seen by the naked eye, such as tiny bloodstains
and sloughed skin cells, can now be revealed through DNA profiling.
Innovative technology now allows DNA results to be obtained from
a single cell (Findlay et al., 1997). Note, however, that normal casework
involves optimal requirements and specific limitations that will be dis-
cussed in Chapter5 covering low levels of DNA.
As of March 2013, the Innocence Project in New York has helped exon-
erate more than 300 people (many of whom were on death row) through
DNA profiling. The leading contributors to these wrongful convictions
were eyewitness identification problems and misleading forensic testi-
mony. Kirk Bloodsworth was the first person sentenced to death whose
conviction was overturned through DNA testing (Innocence Project;
Case 4).
Case 4
Bloodsworth was convicted in 1985 of the rape and brutal kill-
ing of a 9-year-old girl in Maryland in 1984. Although five eye-
witnesses placed him with the girl, he continued to maintain
his innocence. He read about the Pitchfork case in England
while he was in prison and pushed to have the evidence in
his case subjected to DNA testing. Initially, the evidence could
not be located. Eventually the s emen-stained underwear was
found in the judges chambers and the semen DNA profiled.
The DNA did not match Bloodsworths and he was released
from prison in 1993 although not formally exonerated.
Finally, in 2003, prisoner DNA samples were added to state
and federal DNA databases and a match was obtained with
Kimberley Shay Ruffner. A month after the 1984 murder for
which Bloodsworth was convicted, Ruffner was sentenced to
45 years for an unrelated attempted rape/murder and was
incarcerated one floor below Bloodsworth. Ruffner pleaded
guilty to the 1984 murder in 2004.
Through DNA testing, the real perpetrators have been found in nearly
40% of these exonerations pursued by the Innocence Project. Some of this
testing was performed after the original DNA tests were deemed faulty
or the original forensic laboratory mishandled evidence. A perusal of the
website profiles of the exonerations (Innocence Project) makes for infor-
mative reading for a criminal justice professional.
This chapter has covered the continuing growth and strength of
DNA evidence used in criminal cases. The following chapters will
describe the techniques and also the strengths and limitations inherent in
any DNA analysis performed at a crime scene, during a medical examina-
tion, or in a laboratory.
References
Aronson, J.D. 2007. Genetic Witness. New Brunswick, NJ: Rutgers University Press.
Avery, O.T., Macleod, C.T., and McCarty, M. 1944. Studies on the chemical nature
of the substance inducing transformation of Pneumococcal types. Journal of
Experimental Medicine, 79, 137159.
Findlay, I., Taylor, A., Quirke, P. et al. 1997. DNA fingerprinting from single cells.
Nature, 389, 555556.
Franklin, R. and Gosling, R.G. 1953. Molecular configuration in sodium thionucle-
ate. Nature, 171, 740744.
Gill, P., Jeffreys, A.J., and Werrett, D.J. 1985. Forensic application of DNA finger-
prints. Nature, 318, 577579.
Gill, P. and Werrett, D.J. 1987. Exclusion of a man charged with murder by DNA
fingerprinting. Forensic Science International, 35, 145148.
Goldman v. Simpson, Cal. Sup. Ct., LA County, Case SC036340.
Human Genome Project. http://www.ornl.gov/sci/techresources
Innocence Project. http://www.innocenceproject.org
Jeffreys, A.J., Wilson, V., and Thein, S.L. 1985. Individual specific fingerprints of
human DNA. Nature, 316, 7679.
Jeffreys, A.J., Brookfield, J., and Semeonoff, R. 1986. Positive identification of an
immigration test case using DNA fingerprints. Nature, 317, 818819.
Jeffreys, A.J., Allen, M., Hagelberg, E. et al. 1992. Identification of the skeletal
remains of Josef Mengele by DNA analysis. Forensic Science International, 56,
6576.
Mendel, G. 1865. Experiments in plant hybridization. English translation by
C.T. Druery and W. Bateson. Journal of the Royal Horticultural Society, 26, 132,
1901.
National Research Council. 1992. DNA Technology in Forensic Science. Washington:
National Academy Press.
Nature. http://www.nature.com
Orchid Cellmark. http://www.orchidcellmark.ca/site/casefiles
People v. Castro, 545 NYS 2d 985 (Sup. Ct. 1989).
People v. Simpson, Cal. Sup. Ct., LA County, 1995, Case BA097211.
R. v. Doheny and Adams, 1 Criminal Appeals R 369, 1997.
State v. Woodall, 385 SE 2d (W. Va. 1989).
Taupin, J.M. 2001. DNA analysis and drug trafficking. Forensic Bulletin,
November 18 (Australia).
Taupin, J.M. 1994. Impact of DNA Profiling on the Criminal Justice System.
MA thesis, Department of Criminology, University of Melbourne, Australia.
http://www.dtl.unimelb.edu.au/dtl_publish/research/28/66762
Venneman, M. and Koppelkamm, A. 2010. mRNA profiling in forensic genetics I:
possibilities and limitations. Forensic Science International, 203, 7175.
Wambaugh, J. 1989. The Blooding. New York: William Morrow.
Watson, J.D. 1968. The Double Helix. London: Penguin Books.
Watson, J.D. and Crick, F.H.C. 1953. A structure for deoxyribose nucleic acid.
Nature, 171, 737738.
Watson, J.D. and Crick, F.H.C. 1953. Genetical implications of the structure of
deoxyribose nucleic acid. Nature, 171, 964967.
Werrett, D.J., Lygo, J.E., and Sutton, J.G. 1989. The introduction of DNA analysis
into Home Office laboratories in England and Wales, Banbury Report 32.
DNA Technology and Forensic Science, 233240.
Wilkins, M.H.F., Stokes, A.R. and Wilson, H.R. 1953. Molecular structure of deoxy-
pentose nucleic acids. Nature, 171, 738740.
chapter two
Strengths and limitations of

DNA profiling evidence
2.1Introduction: Power and caution
DNA profiling has become a well-k nown technique used in criminal and
other legal cases due to the massive publicity generated by high profile
cases, TV crime shows, and films. DNA profiling is now used by organiza-
tions such as the Innocence Project in New York to rectify past errors in
criminal cases. More than 50% of convictions that project has overturned
were attributed to faulty forensic evidence presented in the original tri-
als (Innocence Project). However, DNA must be present on evidentiary
items for exoneration to be achieved. Exhibits from very old cases may
have been lost, destroyed, or decomposed. Furthermore, no biological evi-
dence may have been obtained in the original trial.
Every legal professional, forensic scientist, and indeed member of the
public should know that DNA profiling is neither infallible nor unassail-
able. Many documented instances demonstrate that contamination, tran-
scription, and other errors have been made in DNA profiling (as described
in Chapter8) simply because the technique is performed and interpreted
by human beings. It is also possible that DNA of an accused found at a
crime scene may have an innocent explanation and/or the DNA may not
be relevant to the crime.
Forensic means pertaining to the courts of law. The concept of science
is less well understood, even by some practicing forensic scientists.
Science is a method of study used to understand and describe the physical
universe around us. The discipline of science is defined by the notion of
hypothesis testing. First a hypothesis or theory is proposed. Experiments
are then performed to test the hypothesis. The results of the experiments
will either support or refute the hypothesis. The scientific method pro-
vides the framework for the testing of hypotheses. Alternative hypotheses
should always be provided (Taupin and Cwiklik, 2010).
Science is a dynamic endeavor whereby new discoveries change the
way we think about the world. This is demonstrated in forensic science
when the discovery of a more discriminating technique is used to distin-
guish samples that previously could not be differentiated. DNA profiling
13
is the best known example of a distinguishing technique. DNA profiling

for use in forensic cases was a significant scientific achievement in the
1980s. It is an extremely powerful tool because it demonstrates very high
discriminating power on samples from different individuals. The high
exclusionary factor should be the focus of our attention, as all forensic
science starts from an exclusionary perspective. Unfortunately the high
statistics associated with the probability of a match have diverted atten-
tion away from exclusion and toward inclusion and matching.
DNA profiling is perceived to be more objective than other foren-
sic methods. Furthermore, it is robust and can be performed on ancient,
poor quality, or very small samples, unlike the traditional techniques. No
doubt these are the reasons behind its very quick uptake by the forensic
and legal community; its notoriety in the solution of some high profile
unsolved crimes also contributed to its widespread use.
Accurate individualization is the aim of all forensic identification sci-
ence. The problem is that the various types of forensic identification sciences
do not use the same scientific paradigms or report their conclusions in
the same format. DNA evidence is probabilistic and quantitative, while
fingerprint evidence is categorical (match or n onmatch). Forensic iden-
tification procedures may lead to categorical elimination, but unless the
number of potential sources is limited and known, no forensic identifica-
tion procedure can lead to categorical identification.
A murder case from Arizona in the United States illustrates the con-
sequences of unreserved acceptance of forensic evidence and represents
a stark contrast between forensic analysis that has a solid scientific foun-
dation and questionable forensic evidence (Innocence Project). Case 1
exemplifies the advantages of applying DNA profiling to old cases adjudi-
cated when routine blood grouping was not discriminatory.
Case 1
The body of a waitress was found on a December morning in
1991 in a bar where she worked. She had been fatally stabbed
and investigators focused on bite marks on her body. Upon
hearing that the victim told a friend that Ray Krone helped her
close the bar the previous night, the detectives asked Krone
to make an impression of his teeth. Due to an accident, Krone
had a distinctive dental pattern; experts stated that it matched
the bite marks on the victims body. Saliva on the victims body
came from someone with the most common ABO blood group,
that of Krone.
During trials in 1992 and 1996, Krone was convicted of
murder and kidnapping, mainly due to bite mark testimony of an
Chapter two: Strengths and limitations of DNA profiling evidence 15
alleged expert. Krone was dubbed the snaggle tooth killer

and sentenced to life in prison.
Then, in 2002, as a result of an investigation by Krones
lawyer, DNA testing was conducted on the saliva found on
the clothing associated with the bite marks on the victim.
DNA profiling showed that the saliva did not match Krones.
A search of the Combined DNA Index System (CODIS) data-
base matched the DNA to that of Kenneth Phillips, an inmate
at an Arizona prison. Krone was released in 2002 and Phillips
entered a guilty plea in 2006. The expert then noted that the
bite marks were not uniquely Krones and were consistent with
the dental pattern of Phillips.
The advent of DNA profiling has no doubt improved the recognition of

biological evidence by investigators and the judicial system and led to
questioning some unfounded assumptions. Case 2, also from the United
States, is an example (Pilkington, 2012).
Case 2
On September 29, 2012, Damon Thibodeaux became the 300th
person to be exonerated by DNA evidence in the United States.
He had been on death row in Louisiana since 1997 for the rape
and murder of his 14-year-old step cousin. He had given a con-
fession but much of what he said did not fit the facts.
No semen was found on the body of the victim. The police
theorized that semen was originally present but maggots
destroyed it. Clothing from both Thibodeaux and the victim
was reanalyzed and no DNA transfer between accused and
victim was found. In fact, no evidence of rape was found and the
maggot theory was absurd.
Unfortunately, a mystique surrounds the utility of DNA profiling. The

general public perception is that a suspect must be guilty of an offense if
DNA evidence is found. This perception can extend to the criminal justice
system and even forensic scientists. Gill and Buckleton (2010) noted in the
forensic literature that this is highly dangerous thinking.
2.2Discrimination power of DNA profiling

The hypothesis testing should be an inherent part of a forensic examina-
tion. Forensic science comes from an exclusionary view. What makes a
forensic test so valuable is its discriminating power, that is, its potential
for excluding. This is one reason why DNA profiling evidence is so pow-
erful. Its discriminating power may be in the order of billions.
Prior to the introduction of DNA profiling, the biological tests used
to analyze forensic samples had very low discriminating power. The
comparison of evidence samples to reference samples is a distinguish-
ing characteristic of forensic science. Previously, routine serology such as
ABO blood grouping was used to attempt to differentiate evidence and
reference samples. Their discrimination levels were poor. For example,
blood group A is common to a third of the population. For this reason
blood grouping and other methods were most often used as exclusion-
ary tools. However, samples taken from the same ABO blood group will
have different DNA profiles and may be differentiated by DNA testing as
shown in Case 1 above.
Typical DNA profiling techniques examine at least 10 different areas
on a DNA molecule and this procedure contributes to the very high
chance probabilities. Like the other biological markers traditionally used,
the advantages of DNA profiling include its genetic basis, backup from
statistics, and availability of databases. The technique can be used to con-
clusively exclude an individual as the donor of an unknown source of
DNA, if the DNA profiles from the individual and an unknown source
are different. If the DNA profiles are the same, the statistical probabil-
ity of obtaining a profile from a random individual in the population
is determined.
Crime stains, however, do not always yield full profiles (they may
be partial) or single profiles (they may be mixed from two or more indi-
viduals) or they may be insufficient. The resulting lower discrimination
and/or evidential value may be crucial in a criminal case and these issues
are discussed in Chapter5.
2.3Genetic basis for DNA profiling

DNA profiling has a solid scientific foundation based on the science of
molecular biology and genetics. The study of genetics and DNA is now
part of high school biology studies and many texts cover the subjects.
DNA profiling for forensic purposes is more specific and references will
be given for each aspect covered in a chapter.
A cell is the smallest working unit of a living organism. It consists of a
liquid called cytoplasm that contains instructions for chemical processes,
and a nucleus surrounded by an external membrane. The DNA molecule
in the nucleus contains the genetic instructions for the development
and functioning of a living organism. Nuclear DNA is the DNA found

within the nucleus of a cell and two types of DNA profiling use this DNA.
Autosomal STR (short tandem repeat) profiling is the most common
DNA typing system used in criminal cases and exploits the differences
found on different chromosomes (see below for explanation of STR).
Y-STR profiling examines only the Y chromosome in the nucleus, is inher-
ited paternally, and has a specific purpose (see Chapter 6). Mitochondrial
DNA (mtDNA) profiling examines DNA in mitochondria, found in the
cytoplasm of the cell. MtDNA is inherited maternally and mtDNA profil-
ing also has a specific purpose (see Chapter 7).
A gene is considered a unit of heredity in a living organism. The loca-
tion of the gene on the chromosome is called the locus and the differ-
ent versions of a gene are called alleles. All humans have the same genes
at the same loci on the same chromosomes, but the alleles differ from
individual to individual. This makes every person unique from a genetic
point of view.
A genotype is the set of alleles of a gene. A homozygous genotype
means that two identical alleles exist at the same locus, that is, the same
allele was inherited from both parents. A heterozygous genotype reveals
two different alleles at the same locus different alleles were inherited
from the organisms mother and father. Generally, a female has the X chro-
mosome only (denoted as X,X) and the male has one X and one Y (denoted
X,Y). A DNA profile is the combination of genotypes obtained for differ-
ent loci. It is important to remember that multiple loci are examined in
DNA profiling to reduce the possibility of a coincidental match of unre-
lated individuals.
Specific areas on the DNA molecule that are known to vary widely
among individuals are examined using specific technical kits. The areas
vary in length among different people and are called short tandem repeats
(STRs). These repeat units are also called microsatellites and vary from 2 to
5 base pairs, typically repeated 5 to 30 times (Tamakin and Jeffreys, 2005).
The STRs used in DNA profiling kits cover four base pairs. Analyzing the
number of repeat units at many loci provides a highly sensitive measure
of individual identity. STR markers are typically chosen to avoid problems
of linkages between markers (see Chapter 4 on statistics).
The STR markers need to be the same for DNA profile comparison
within jurisdictions and across countries; however currently there is no
one uniform marker set used although a set of core STR loci is used in
Europe and the USA (see Section 2.7). The Amelogenin marker is the com-
mon sexing marker that determines whether X or Y or both is present.
There are separate criteria for determining the requirements for STRs
in criminal casework and for database intelligence. Expanding a set of
core STR loci: (a) reduces the likelihood of adventitious matches in the
database; (b) increases international compatibility to assist law enforce-
ment data sharing; and (c) increases the discrimination power to assist
missing person cases (Butler and Hill, 2012) and thus is database focused.
Criminal casework considerations require a focus on sensitivity, due to
crime DNA samples often being compromised in some way (low level,
degraded, or otherwise sub-optimal).
2.4Stability of DNA profiling

Routine serology markers used in forensic science before the advent of
DNA profiling, such as ABO grouping or typing of enzymes, were often
not applicable because the biological samples were small or the enzymes
degraded quickly. The polymerase chain reaction (PCR) was a boon for
forensic science, as it enables the analysis of minute quantities of blood
and semen and is effective for degraded samples such as those commonly
encountered at crime scenes. PCR is essentially a molecular photocopier
that can amplify very small samples and allow them to be detected
and analyzed.
As biological samples age, the chemicals they contain begin to break
down or degrade. Thus DNA will degrade, albeit more slowly than rou-
tine serology markers. In a biological sample, the interaction of the DNA
and its environment is the factor that determines the preservation of the
DNA. Age is not the only factor. Samples from a few years old to many
decades old may still be analyzed successfully.
The degradation process occurs minimally if the samples are pre-
served (kept dry and free from bacterial attack), but samples degrade
rapidly when exposed to heat, moisture, or sunlight. It is thus impor-
tant to ensure the proper collection and storage of biological evidence.
Biological evidence has been found in improper storage facilities such
as police locker rooms and courtrooms. Efficient storage of DNA extracts
is also important. Reduction in DNA recovery has been observed after
refrigeration of liquid DNA extracts and exposure to multiple freeze
thaw processes.
The stability of DNA over time has been instrumental in solving cold
cases and acquitting the wrongly convicted. Evidence from crime exhib-
its has been recovered and DNA profiled many years after offenses were
committed. Case 3 involving Robert Dewey who was cleared of rape and
murder in May 2012 is an example (Innocence Project).
Case 3
Robert Dewey was sentenced to life in prison in Colorado
in 1994 for the rape and murder of a young woman found in
her bathtub, partially clothed and strangled with a dog leash.
Samples of blood found on Deweys work shirt were said to
be a mixture of biological materials, some of which may have
come from the victim. The bloody shirt and semen-stained
blanket found on the victims couch were stored by the police
in an environment that prevented degradation.
The items were retested with other evidence in 2011. The
DNA from the blood on the work shirt was found to be Deweys.
DNA from semen found on the victims blanket linked (via the
national CODIS database) to Douglas Thames, then in prison
for the rape and murder of another woman in 1989. Partial
DNA profiles from materials under the victims nails and on
the leash used to strangle her also matched Thamess DNA.
It is the opinion of this author that the transfer of DNA onto the body
of a victim, especially in a rape and murder case involving transfer of
appreciable amounts of DNA in semen and saliva, is a crucial factor in
the solution of many cold cases. The clothing and medical exhibits from
victims are often stored indefinitely and DNA profiling can assist in the
resolution of cases many years after the crimes were committed.
Human hairs, particularly the shafts, are less subject to degradation
than blood or semen due to their construction and composition and may
remain intact for centuries (Taupin, 2004). Items from exhibits thought to
have been destroyed may still remain in laboratories or police files. Hairs
on a microscope slide from a robbery were found in police rooms in the
case of Sedrick Courtney after he was in prison for 15 years. Mitochondrial
DNA showed he could not have been the offender (see Chapter7 for case
discussion). It is always worthwhile to investigate further if information
indicates that original police evidence such as clothing from a victim has
been destroyed. Medical swabs or debris collected from the items may
still remain in case files, exhibit rooms, or refrigerators. The stability of
nuclear and mitochondrial DNA makes it possible to obtain a result of
evidential value many years after the commission of a crime.
2.5Persuasive statistics
The ultimate power of DNA profiling is its statistical discrimination. It is
not possible to profile every person in a country, so a statistical probability
is performed if a match is found. The probability statistic is based on the

frequency of each short tandem repeat (STR) at nine or more areas, and
involves multiplication that gives rise to the high values. These values are in
contrast to the relatively low values obtained by traditional serology tests.
Relatedness of the population, sampling effects, racial databases, and
validation studies are necessary before a statistic can be accepted. A prob-
ability of the rarity of the DNA profile or likelihood ratio is produced. A
likelihood ratio compares the prosecution hypothesis against the defense
hypothesis in statistical terms. See Chapter4 for a more detailed discussion.
An accused person may indeed be innocent and query the findings.
One enterprising individual from the Northern Territory of Australia did
so when told of a scientists opinion that the chance of a second person
having the same DNA profile was about one in 50 million. The individ-
ual wanted to know what the police were doing about finding the second
person (personal communication, George Georgiou, Victoria barrister).
Case 4 illustrates the persuasive power of DNA statistics. The miscar-
riage of justice involved Farah Jama in Victoria, Australia and contamina-
tion was a factor.
Case 4
The Victorian Parliament tabled a report about the 2008 convic-
tion of Farah Jama for the rape of an unconscious woman in a
nightclub (Vincent, 2010). Jamas conviction was overturned on
appeal in December 2009; he was initially sentenced to six years
in jail and served 15 months. The inquiry found that DNA evi-
dence was the only link between Jama and the woman. The
report found that the offense probably never occurred and that
medical samples from an unrelated sexual incident with another
woman involving Jama (in which no charges were filed) were
taken by the same medical officer at the same location within
30 hours of taking samples from the alleged rape victim.
Jamas DNA allegedly matched via a database hit when the
samples arrived at the laboratory in an unknown offender
case. Most likely, contamination between the evidentiary sam-
ples occurred during the medical examination, although the
exact mechanism could not be determined. The recommen-
dations of the report included education of legal practitioners
and members of the judiciary on the nature and appropriate
use of DNA.
Justice Vincent stated in the report that the DNA evidence
was perceived to appear so powerful by all involved in the
case that none of the filters on which our criminal justice sys-
tem depends to minimize the risk of a miscarriage of justice
operated effectively until weeks before Jamas appeal. Vincent
noted that no one appeared to be aware of the dangers of rely-
ing on statistical probabilities in the determination of guilt. A
forensic scientist concluded that it was 800 billion times more
likely that the DNA originated from Jama than from another
Caucasian in the Australian population selected at random.
Since the world population is only about 6 billion people,
the 800 billion statistic appeared definitive. Jama subsequently
received monetary compensation from the government.
2.6Relatives
Statistical analyses for DNA profiles are generally quoted for a popula-
tion of unrelated individuals. When considering related individuals in
a particular case, however, the statistics must be reviewed. For example,
the offspring of a parent inherited one allele identical by descent, at each
area profiled on a DNA molecule. The likelihood of observing a DNA
profile in a related individual is thus higher than in the general popula-
tion. Identical twins are particularly problematic, as they develop from
the same egg and thus have the same DNA, as illustrated by Case 5.
Case 5
In January 2009, jewelry valued at $6.8 million was auda-
ciously snatched from a luxury department store in Berlin
(Himmelreich, 2009). Three masked and gloved thieves were
caught on a surveillance camera as they slid down ropes hung
from the stores skylights, thus outsmarting a sophisticated
security system.
DNA was found on a latex glove left at the scene, run
through the national DNA database, and yielded matches
with two peopleidentical twins. They were charged but
released before trial because the court determined that at least
one of the brothers was responsible but could not determine
which one.
See Chapters 4 and 5 for more detailed discussions of statistics.

2.7DNA databases
The United Kingdom became the first country in the world to launch a
national DNA database in April 1995. New Zealand followed closely in
1996. The FBI implemented the Combined DNA Index System (CODIS)
in the United States in 1998. In 2000, the CrimTrac National Criminal
Investigation Database (NCIDD) was launched in Australia. However, not
until 2009 did all jurisdictions in Australia use a single database.
Today, national and regional DNA databases are in use through
statute in many countries including those cited above. These databases
have provided many cold hits for unknown samples left at crime scenes.
Thus, in addition to comparisons with reference samples, biological evi-
dence samples may also be compared to DNA databases to find potential
matches with convicted offenders.
To enable comparisons of DNA profiles both within and between
jurisdictions, the same sets of loci must be analyzed. In 1997, the FBI
selected a core set of 13 STR loci for use within the United States to upload
DNA profiles to the national DNA database. Only 8 of these loci over-
lap with STR data gathered in the UK and most other European nations
(Butler and Hill, 2012). Currently there is a 10 locus system used in the
United Kingdom (AmpFLSTR SGM Plus), 16 locus systems AmpFLSTR
Identifiler and PowerPlex 16 in the United States, and a 9 locus system
AmpFLSTR Profiler Plus in Australia. Efforts are in progress to unify
and expand the number of core STR loci for not only comparison across
jurisdictions but to improve difficult database searches such as familial
DNA where even 16 loci may not be sufficient (see Chapter 7). Profiling
kits are being developed in response to a recommendation to expand
the CODIS core loci from 13 STRs to 20 required loci and 3 optional loci
(Hares, 2012).
DNA databases generally comprise two groups of information: (1)
DNA profiles of convicted offenders and volunteer samples, and (2) DNA
profiles from criminal investigations (crime scene samples) stored in elec-
tronic format. For cases with no suspects, the crime scene DNA profile is
loaded into the database to match against other profiles in specific catego-
ries: (prisoner, suspect, crime scene, limited purpose volunteer, unlimited
purpose volunteer, missing person, and unknown deceased).
Exclusions can help exonerate suspects in an investigation. A link pro-
vides the police with intelligence to assist in their investigations.
The probability of identifying a suspect when a crime scene profile is
checked on the U.K. database is greater than 40% (NDNA, 2009). The U.K.
database contains about 5 million samples (about 10% of the population).
This enormous number is a result of the relatively liberal criteria for entry
into the U.K. database compared to other national databases.
Database matches are mainly through criminal offenses. However, a

DNA database match or link does not represent an arrest or conviction; it
merely indicates that a person may have been at a crime scene.
Cases 1 and 3 from the United States, described above, show the out-
standing utility of DNA databases in identifying the real perpetrators of
violent crimes. Case 4 from Australia illustrates that matches may not
be genuine.
There has been an unacceptable rate of adventitious matches in data-
base searches due to an increasing number of partial profiles in databases
(Schneider, 2009). This is why there have been recommendations to incor-
porate mini STRs as part of the core loci in testing kits so that degraded
or otherwise compromised samples may be more fully analyzed (see
Chapter 5 for explanation of partial DNA profiles).
2.8DNA intelligence-led policing

DNA is now a key component in intelligence-led policing through:
Mass screening of suspects (such as in Case 1, Chapter1*)

Targeting certain crimes in specific areas and collecting DNA
evidence in an effort to catch repeat offenders; this is known as a
targeted operation
National and international DNA databases
Familial matching
Mass screening or DNA dragnets triggered privacy concerns in the

United States and are not considered as productive as such operations
conducted in Europe (Butler, 2012).
There are three different types of DNA database comparisons. The
first is the most common in criminal cases and is a comparison of the
crime scene profile to the suspects profile, and would be accepted as the
random match probability (say p). Database trawls are the second
type, where the crime scene profile is searched for a match against n
number of profiles in the database. There would be approximately np
matches where no one in the database is the source and there is no relat-
edness in the database. The third, which is markedly different and is not
used in criminal investigations, is the trawl through all possible pairs
in a database so that every profile in the database is compared with every
other profile and will result in greater numbers of matches (Kaye, 2009).
The confusion in understanding the third type of database comparison is
* The Pitchfork case was the subject of Joseph Wambaughs 1989 book titled The Blooding.
a mathematical one based on the birthday problem, which is discussed

in Section 4.3.4.
The concept of familial matching is also controversial. A DNA pro-
file from a crime scene may be searched against the DNA database of
a country but only partial components of the profile may be used. This
is because individuals in databases may have close relatives who have
similar DNA profiles that potentially match DNA found at crime scenes.
Chapter7 discusses this topic.
2.9Mass disasters
Catastrophic events involving multiple victims and deaths are termed
mass disasters. DNA profiling has proven invaluable in identifying vic-
tims of mass disasters such as the Boxing Day tsunami in 2004, one of the
deadliest natural disasters in history. The epicenter of this earthquake in
the Indian Ocean was off the coast of Sumatra in Indonesia, and affected
most land masses bordering the Indian Ocean including Indonesia
and Thailand.
Countries such as Australia provided support that included DNA pro-
filing identification of the victims. Smaller scale disasters such as air bal-
looning accidents may require DNA profiling for identification. Remains
of the victims may be comingled, and identification in such instance poses
particular problems. If bodies are not recovered quickly they degrade,
especially in hot and humid climates. Bones or teeth may need to be con-
sidered for DNA analysis. If there is too much degradation, mitochon-
drial analysis and other techniques for analyzing ancient DNA may
be required. A combination of DNA profiling techniques can be used
to achieve positive identification, for example, of the last Romanov tsar,
Nicholas II, and his family (see Chapter7 for a discussion of this fascinat-
ing quest for identification).
2.10DNA evidence in context

The criminal justice professional should consider DNA results in the con-
text of the case and ask several questions. How relevant is the DNA evi-
dence? What is the quality of the other evidence? Is the DNA result the
only evidence of substance? How many DNA results are there? Chapter9
provides a list of questions to consider in determining whether to use
DNA evidence.
Sometimes no DNA results are generated. How do we interpret this?
One possibility is that no DNA was transferred or deposited. This does
not mean that the offender was not at the scene of the crime. The offender
may not have deposited or transferred detectable DNA. Principles of trace

evidence transfer apply (see Section 2.11).
Is DNA always paramount or relevant? Examining an assault case
would require a search for blood on a suspects clothing, because blood
evidence denotes injury to a victim. An examiner should also consider
other evidence types according to observations and hypotheses. A DNA
profile of blood on the suspects clothing that matches the DNA of the
victim may support the presence of the suspect at the scene. However, if
a suspect says he merely helped the victim after an attack by another per-
son, the blood staining on the clothing of the suspect establishes his pres-
ence at the scene but is irrelevant in addressing the assault. The bloodstain
patterns on the clothing may help disprove or support the hypothesis
of the assault and the question of whether the suspect assisted or attacked
the victim.
Similarly, the presence of DNA from semen on a victim that matches
the DNA profile of the accused is not the paramount factor in a case of
rape fought on consent grounds. DNA can be obtained from handled
objects, but the association of material that cannot be sourced to a body
fluid such as blood or semen reduces the relevance of the DNA profile.
2.11Time of deposition: Transfer

and persistence of DNA
DNA analysis has now advanced so that a profile can be produced from
a single cell (Findley et. al, 1997). This means that we can analyze DNA
that is not visible to the naked eye, and must consider issues such as the
type of the material (blood, semen, skin cells), how the DNA may have
been transferred, and how long it has been present at a crime scene or on
an exhibit.
The principle of trace evidence transfer is Locards Theorem, which
is essentially every contact leaves a trace. This trace may or may not
be detectable according to the type and amount of material in ques-
tion and the method used. The principle is the basis for examination in
many forensic textbooks to explain the rationale for the testing of exhibits
(e.g., Taupin and Cwiklik, 2010).
The potential for secondary transfer was first described in Melbourne,
Australia. A study showed that plastic tubes held for fixed times by dif-
ferent consecutive users produced the DNA profile of the last holder and
sometimes evidence of the DNA profile of the previous holder (van Oorshot
and Jones, 1997). The authors warned of the potential for secondary trans-
fer and contamination.
Primary transfer occurs when DNA is transferred from a person to

an item. Secondary transfer is the transmission of that DNA to another
item. Tertiary transfer occurs when the DNA on the second item is in
turn transferred to a third. Many published studies have noted second-
ary transfers in quantities that can be detected from items that people
simply touched to other items (Wickenheiser, 2002).
DNA alone cannot be related to a specific action. DNA obtained from
an object may have been deposited at separate times and places unrelated
to the event in question if a body location cannot be determined. DNA
from an unspecified cellular source reduces the relevance of biological
evidence to a particular event and increases the uncertainty as to how
the DNA may have been transferred to the item. Thus the relevance of
findings may be difficult to assess (Gill and Buckleton, 2010). Chapter3
provides more discussion on this topic.
Case 6 comes from the authors files and illustrates that the deposition
of DNA does not necessarily tie an accused to a crime event.
Case 6
An armed robbery trial was held in 2012 in the outback of New
South Wales, Australia. The accused was an indigenous male.
The trial was the second; the first trial ended in a hung jury.
It was alleged that the accused threatened a shop attendant
in a convenience store with a knife and a wrench, demanded
cash, and left with money from the till. The attendant said the
offender asked for and obtained a drink of water from a cooler
inside the store. The offender then left two plastic disposable
cups on the counter, one inside the other. Only one squashed
cup was received by the laboratory.
The top centimeter of the inner and outer rims of the cup
was sampled. A full single source DNA profile matching that
of the accused was obtained. The body origin of the DNA
was not determined and was assumed by the laboratory to be
saliva. The likelihood that the DNA profile would match an
unrelated individual in the New South Wales population was
estimated at 1 in 333 billion. However, statistics were not given
for the indigenous population of the state (known to be more
closely related than nonindigenous residents).
Furthermore, only one cup was received by the laboratory,
whereas two cups were left on the counter and examined by
crime scene personnel. There was confusion about where the
single cup originated. Empty plastic cups stacked on the water
cooler inside the store appeared to be both used and unused.
The uncertainty as to which cup was used by the offender and

at what time (not necessarily at the time of the offense), and a
DNA match statistic not relating to the indigenous origin of the
accused, led to a not guilty jury verdict.
The number and type of sample impacts the evidential value of DNA
in a case. A two-way transfer with two separate DNA profiles (DNA in
blood found on a suspect matches DNA of the victim and semen DNA on
the victim matches DNA of the suspect, for example) is far more powerful
than one DNA profile from an unspecified body origin.
2.12Relevant evidence?
It may not be possible to determine the time of deposition of DNA, so
the DNA evidence may become less probative. How the DNA got on
the exhibit and whether the DNA is relevant to the offense may also be
paramount questions. The murder of Meredith Kercher in Italy (Case 7)
was a high profile case covered by media around the world. The trial and
appeal involved these factors in an interesting manner for criminal justice
professionals (Hellmann, 2011). Amanda Knox, one of the three accused
individuals, was from the United States. Raffaele Sollecito, her then boy-
friend, was from Italy. The third individual was the Ivory Coast-born
Rudy Guede. Knox and Sollecito were freed on appeal in 2011 (Hanlon,
2011). The original verdict was criticized by one of the appeal court judges
(Kington, 2011). In March 2013 Italys highest appeal court quashed the
acquittals and ordered a fresh trial due to the manner in which the appel-
late court had been conducted (Davies, 2013).
Case 7
A British exchange student, 21-year-old Meredith Kercher, was
stabbed to death in Perugia, Italy in 2007. She was found in her
bedroom in an apartment she shared with three other females
including Amanda Knox. Kercher was found on the floor with
stab wounds to her throat; she was sexually assaulted and
some of her belongings had been stolen. Rudy Guede was con-
victed in 2008 of murdering and sexually assaulting Kercher
and sentenced to 30 years, reduced to 16 years on appeal in
December 2009.
The evidence against Guede appeared uncontroversial,
since DNA profiles matching his were found on Kerchers
body and clothing (Hellmann, 2011). The key DNA evidence
against Knox and Sollecito was on bra clasps from the victim
at the crime scene and on a knife found in a kitchen drawer at
Sollecitos flat.
Knox was sentenced to 26 years and Sollecito to 25 years in
December 2009. The conviction was quashed in 2011 through
a successful appeal. The evidence of DNA defense experts
Carla Vecchiotti and Stefano Conti was crucial to the appeal
(Conti-Vecchiotti, 2011). The knife allegedly had traces of DNA
matching Knox on the handle and DNA matching Kercher on
the blade. The DNA alleged to have come from Knox was not
disputed (it was found in her boyfriends flat) but the DNA
profile alleged to have come from Kercher was very low level
DNA (see Chapter 5 for a discussion on low level DNA and
this case). There was no evidence that this low level profile
was from blood. The suspects and victim knew each other and
had access to each others apartments. It was not obvious why
the knife was believed to be evidential. Questions were raised
about handling and packaging of evidence.
The bra of the victim had been cut and was found at her
feet. The bra clasp was a fragment of material with a deformed
clasp that had been removed from the rest of the bra and origi-
nally observed under Kerchers body. The bra clasp displayed
a clear major/minor mixture profile (see Chapter5 for mixture
discussion). There was no dispute that the major DNA pro-
file on the clasp came from Kercher (she had worn the bra).
The minor DNA component was alleged to have come from
Sollecito. Y
-STR analysis (see Chapter 6 for discussion of the
technique) showed a mixture from at least three males. The
bra clasp was collected under a mat on the floor, more than a
meter from its original position. It was also collected 46 days
after the crime in a context highly suggestive of environmen-
tal contamination.
The appeal panel consisted of six Italian citizens and two
judges. The court ordered a review of the DNA evidence; the
judges wrote that originally the scientific investigations occu-
pied a preeminent position (Hellmann and Zanetti, 2011). The
appeal court could not rule out contamination for the knife or
the bra clasp and stated that low level DNA precautions did not
appear to have been applied for the knife DNA. The court also
noted an erroneous interpretation of both the autosomal DNA
and Y-STR profiles on the bra clasp.
This case is illustrative of the prosecution inferring the association

of an activity such as stabbing or handling a bra clasp with a DNA profile
that could not be sourced to a particular time or body fluid (similar to
Case 6 above). Furthermore, collection procedures were issues in the deci-
sion. This case demonstrates the many factors that must be considered in
the collection, handling, and interpretation of DNA evidence, particularly
testing of low level or small amounts of DNA. When a DNA profile can-
not be associated to a body fluid such as blood or semen, and collection
practices are questionable (bra clasp collected 46 days after the crime),
the criminal justice professional, whether acting for the prosecution or the
defense, must evaluate the probative value of the DNA. Chapter8 further
discusses this case in relation to contamination and collection issues.
Chapter5 explains low level DNA techniques.
2.13Relevant exhibits?
The DNA evidence from the exhibits in a criminal case should accord
with the hypotheses proposed, unless there is a plausible explanation.
Sometimes the wrong exhibits are examined or the results may be insuf-
ficient. Case 8 illustrates the failure to consider all items submitted to a
laboratory in the context of a case (Queensland Court of Criminal Appeal,
2001; Taupin and Cwiklik, 2010).
Case 8
A 13-year-old girl was raped in Australia in 1999. She named
Frank Button as the offender. Spermatozoa were obtained from
vaginal swabs but no DNA profile could be obtained. Sheets
and pillowcases from the girls bedding were also sent to the
laboratory but not tested. Button was convicted and sentenced
to seven years imprisonment.
His appeal was heard on the grounds that no scientific evi-
dence was presented at the trial. The laboratory then tested the
bedding on insistence from the defense lawyers. The DNA pro-
file from semen on the bedding did not match Buttons DNA
profile. The laboratory then retested the vaginal swabs and
obtained a DNA profile that also failed to match Buttons. The
vaginal swab profile matched that of a convicted rapist and
Button was acquitted.
The criminal justice professional should be alert to the possibility of other

exhibits that may have been collected by the police, may not have been
tested forensically, and may be probative.
2.14CSI effect and the notion of infallible

forensic evidence
The CSI effect is named after the eponymous television show, one of the
worlds most popular TV shows that has spawned many sequels and imita-
tors. These shows portray forensic scientists and crime scene investigators
as clever and morally correct individuals who fight to clear the names of the
innocent and put real criminals behind bars. The techniques portrayed on
these shows also lead viewers to believe that they are viewing forensic sci-
ence at work and that the science is infallible. However, the technology por-
trayed is intended for entertainment purposes instead of scientific accuracy.
The worldwide legal fraternity believes the CSI effect has changed
the way many trials are presented in that prosecutors are pressured to
deliver more forensic evidence in court. Juries query the absence of foren-
sic evidence and are likely to give more credence to prosecution cases
that contain it. Defense lawyers say that the CSI shows make juries more
unwilling to see that scientific findings can be compromised by human or
technical errors. Although highly publicized, the reality of the CSI effect
remains uncertain and we have no real evidence to conclude that jurors
verdicts are distorted by it (Goodman-Delahunty and Verbrugge, 2010).
2.15Relationships of lawyers and scientists

Scientists and lawyers have a constrained relationship in the courtroom.
Scientists may find it quite difficult to convey complex scientific principles
in lay terms to lawyers and to juries. Defense lawyers may think they are
opening Pandoras box if they challenge the forensic evidence, for fear
that it may appear even more infallible.
There is a challenge in communicating successfully the true probative
potential of DNA evidence to trial judges and jurors. Lay jurors are likely
to need some guidance in making sense of evidence expressed in terms
of probabilities. Chapter9 provides guidelines for criminal justice profes-
sionals considering DNA evidence, along with a list of questions to ask
themselves and their expert DNA witnesses.
References
Butler, J. 2012. Advanced Topics in Forensic DNA Typing: Methodology. San Diego:
Elsevier Academic Press.
Butler, J. and Hill, C. 2012, Biology and genetics of new autosomal STR loci useful
for forensic analysis, Forensic Science Review, 24, 1526.
Davies, L. 2013. Amanda Knox and Raffaele Sollecito face retrial over Meredith
Kercher murder. The Guardian, March 26. http://www.guardian.co.uk/
world/2013/mar/26/amanda-knox-retrial-meredith-kercher-murder.
Findlay, I., Taylor, A., Quirke, P. et al. 1997. DNA fingerprinting from single cells.
Nature, 389, 555556.
Gill, P. and Buckleton, J. 2010. A universal strategy to interpret DNA profiles
that does not require a definition of low copy number. Forensic Science
International: Genetics, 4, 221227.
Goodman-Delahunty, J. and Verbrugge, H. 2010. Reality, fantasy, and the truth
about CSI effects. InPsych, August. http://www.psychology.org.au/
publications/inpsych/2010/august/goodman
Hanlon, M. 2011. As Amanda Knox walks free, now DNA evidence is on trial.
Daily Mail Online, October 5. http://www.dailymail.co.uk/debate/article-
2044935/Amanda-Knox-freed-Now-DNA-evidence-trial-Kercher-murder-
acquittal.html
Hares, DR. 2012. Expanding the CODIS core loci in the United States. Forensic
Science International Genetics, 6, e52.
Hellmann, P. 2011, The Hellmann-Zanetti Report on the Acquittal of Amanda Knox and
Raffaele Sollecito. http://hellmannreport.wordpress.com
Himmelreich, C. 2009. Despite DNA evidence, twins charged in heist go free. Times
Online. http://www.time.com/time/world/article/08599,1887111,00.html
Kaye, D. 2009. Trawling DNA databases for partial matches: What is the FBI afraid
of? Cornell Journal of Law and Public Policy, 19, 145171.
Kington, T. 2011. Amanda Knox trial was flawed at every turn, says appeal judge.
The Guardian, December 15.
National DNA Database, United Kingdom. http://www.homeoffice.gov.uk/
science-research/using-science/dna-database
Pilkington, E. 2012. Louisiana death row inmate freed after 15 yearswith a little
help from DNA. The Guardian, December 7.
Queen v. Frank Allan Button. 2001. Queensland Court of Criminal Appeal, QCA 133.
Schneider. 2009. Expansion of the European Standard Set of DNA database loci
the current situation. Profiles in DNA, March. http//promega.com
Tamaki, J. and Jeffreys, A.J. 2005. Human tandem repeat sequences in forensic
DNA typing. Journal of Legal Medicine, 7, 244250.
Taupin, J.M. 2004. Forensic hair morphology comparison: A dying art or junk sci-
ence? Science and Justice, 44, 95100.
Taupin, J.M. and Cwiklik, C. 2010. Scientific Protocols for the Forensic Examination of
Clothing. Boca Raton, FL: CRC Press.
van Oorshot, R.A. and Jones, M. 1997. DNA fingerprints from fingerprints. Nature,
387, 767.
Vecchiotti, C. and Conti, S. 2011. Conti-Vecchiotti Report. http://knoxdnareport.
wordpress.com
Vincent, Justice F. 2010. Inquiry into the Circumstances that Led to the Conviction
of Mr. Farah Abdulkadir Jama. Victorian Government Printer, Melbourne,
Australia, 2010.
Wambaugh, J. 1989. The Blooding. New York: William Morrow.
Wickenheiser, R.A. 2002. Trace DNA: A review, discussion of theory, and applica-
tion of the transfer of trace quantities of DNA through skin contact. Journal of
Forensic Sciences, 47, 442450.
chapter three
DNA profiling basics
3.1What is DNA?
Deoxyribonucleic acid (DNA) is a complex chemical found in the nuclei
of all cells of the human body, except red blood cells. It is considered a
genetic blueprint that is responsible for our chemical and physical charac-
teristics. Half of an organisms DNA is inherited from each parenthalf
from the mothers egg that is fertilized and half from the fathers sperma-
tozoa. Watson and Crick were awarded the Nobel Prize in 1962 for their
discovery of the double helix structure of the DNA molecule. Their work
was published in Nature (Watson and Crick, 1953). See Chapter1 for a dis-
cussion of the history of DNA profiling.
Each persons DNA remains the same over his or her lifetime and the
composition of the molecule remains the same throughout the body. This
is a forensic advantage, because the DNA from a bloodstain at a crime
scene can be compared with DNA from a reference saliva swab from a
victim or suspect.
Traditionally biological fluid typing, known as serology, was used as
an investigative technique for solving violent crimes because biological
materials are shed during violent acts. For example, blood is commonly
found at homicide scenes and semen is found in rape cases. Today, DNA
is more discriminating than traditional serology testing and it can be
obtained from materials even when it cannot be seen.
3.2Biological materials allowing DNA profiling

Physiological fluids and biological materials are the most common types
of physical evidence found in violent crime cases. The advent of DNA typ-
ing for individualization has increased its importance. Before the intro-
duction of DNA profiling, the biological tests used to analyze forensic
samples had very low discriminating power. Biological fluids such as
blood, semen, and saliva were tested using ABO grouping and enzyme
and protein tests that needed reasonable (visible) sample quantities and
had low discrimination power. For example, blood group A is shared by
one-third of the population.
33
Specific separate confirmatory tests are required to confirm the pres-

ence of blood, semen, or saliva. A DNA profile is specific to human mate-
rial. However, to identify a body (somatic) location, a confirmatory test
(such as the presence of spermatozoa to confirm semen) is required to
associate a DNA profile to a biological material. The testing becomes even
more complex if a sample contains a mixture of DNA profiles and poten-
tially a mixture of biological fluids. A blood stain with a DNA profile
mixture of at least two contributors must be considered in context. Could
semen be present as well? Research is in progress to identify somatic ori-
gins from mixed samples and mixed DNA profiles.
Blood and semen can be more readily associated with crimes due to
their associations with violent or intimate contacts. However, trace DNA
from an unspecified cellular source reduces the relevance of such biologi-
cal evidence to a crime.
Touch DNA is defined as arising from cellular materials in nucleated
epithelial cells from the skin surfaces. The specific body location and time
of deposition cannot be determined. Touch DNA may be obtained from
handled objects but its presence is dependent on the amount of handling
and the sheddability of the skin cells of the handler. Sheddability may be
dependent on environmental conditions. DNA can be obtained from a wide
variety of handled objects such as steering wheels, pens, cards, and bags,
but the nature of these objects means that DNA found on them may have
come from multiple contributors and thus produce mixed DNA profiles.
The association of a specific body fluid and a DNA profile is not
implicit (Taupin and Cwiklik, 2010; Peel and Gill, 2004). Currently DNA
profiling can identify an individual from a sample of biological material
but it does not reveal the body fluid or tissue source from which the profile
originated. The determination of the type of body fluid is important both
for evidential value and also to ensure the correct handling of samples.
The amount of DNA borne per volume of sample material or exhibit
varies according to the source. Solid human tissue and sperm samples of
DNA contain very large amounts of DNA per unit volume. These samples
should be considered as having the highest DNA potential when two or
more types of DNA-bearing cells are found on one item (Wickenheiser,
2002). Blood has the next highest DNA potential. Although blood is com-
monly found in violent crime cases, the DNA-bearing white blood cells are
outnumbered 400 to 1 by red blood cells. Saliva and nose and mouth secre-
tions exhibit the third-highest DNA potential because of the small volume
of body fluid conveying DNA-bearing cells and small contact areas.
Wearer DNA may be found on clothing but the quantity depends on the
DNA deposited and the time elapsed. A garment may act like a reservoir
of DNA if washed infrequently. Because of its transient nature, trace DNA
Chapter three: DNA profiling basics 35
Human tissue
Sperm
Blood
Saliva
Wearer DNA
Touch DNA
Figure 3.1 Relative DNA contents of biological materials found on crime exhibits.
has the lowest DNA potential. Figure 3.1 shows general order of DNA
presence in human biological fluids.
3.2.1Searching for DNA on exhibits

The examination of biological evidence such as blood illustrates the steps
in a forensic examination. The main objectives of biological evidence anal-
ysis are identification (or classification), individualization (DNA typing),
and reconstruction. The blood must first be located. This may be a diffi-
cult task if it is deposited in minute quantities or on dark colored surfaces.
Presumptive and/or confirmatory tests are then performed to identify
stains or deposits such as blood that the examiner thinks may be of interest.
Sampling will be an issue whenever an exhibit reveals multiple blood
or other biological stains. How many stains should be sampled and how
many tests should be performed? A DNA analysis result may be detri-
mental if all material is consumed in confirmatory tests. In any criminal
case, the decision to sample or perform a particular test should be based
on background information about the alleged crime and the scientific
method used to test the hypotheses.
The first step in identifying a body fluid is highly important since
the nature of the fluid is itself very informative to the investigation.
Furthermore, the destructive potential of a screening test must be con-
sidered when only a small amount of material is available. The ability to
characterize an unknown stain at the scene of a crime without having
to wait for results from a laboratory is another critical step in forensic
body fluid analysis.
Current tests for the identification of body fluids use chemolumines-
cence and the detection of specific proteins (Verkler and Lednev, 2009).
Significant advances in laser technology and the development of novel
light detectors have dramatically improved spectroscopic methods for
molecular characterization over the past decade. Because gene expres-
sion patterns are tissue specific, a determination of the type of body fluid
based on messenger RNA (mRNA) profiling may eventually be possible
in routine case work. RNA can be isolated in suitable quality and quan-
tity from blood, saliva, and vaginal secretions. A system using mRNA
that can identify blood, saliva, semen, and menstrual blood in individual
stains or in mixtures of body fluids has been developed (Fleming and
Harbison, 2010).
It should be remembered that blood stains may mask semen stains,
and DNA from a sample may come from a number of sources. Mixtures
from different body fluid sources and/or individuals may appear on one
exhibit or in a single stain (Taupin and Cwiklik, 2010). A mixed DNA pro-
file from two individuals and two possible body sources may result. The
analysis of semen on a complainants bed sheet may yield a mixed DNA
profile. A major DNA component (or sperm fraction from differential
lysis) of a stain likely to have come from semen may correspond to the
DNA profile of a suspect.
A minor DNA component likely to have come from epithelial cells
may correspond to the DNA profile of a victim. It is not possible in these
situations to determine whether the mixed stain occurred from biological
fluids transmitted during sexual intercourse (Petricevic et al., 2006). Nor
is it possible to determine whether the DNAs from multiple individuals

were deposited at the same time.
3.2.2Blood
Blood has traditionally been an excellent source of DNA (from the white
blood cells). The quantity required for DNA analysis has decreased over
the years as the technique has become more sensitive. Stains of 1 mm size
or smaller can be analyzed as shown by Case 1.
Case 1
A young boy named Damilola Taylor died in 2000 on a London
housing estate as a result of a stab wound to his thigh that
caused extensive blood loss at the scene (Rawley and Caddy,
2007). No forensic evidence was presented during the first trial
of four boys in 2002. Two were found not guilty and charges
were dropped against the other two.
During a second police investigation, clothing belonging
to two brothers, Danny and Ricky Preddie, were submitted
for re-examination at a different forensic laboratory from the
one that initially examined more than 400 clothing items. A
small drop of blood found on the heel of a white training shoe
belonging to Danny Preddie was DNA profiled and found to
match the profile of Damilola Taylor. A bloodstain was also
found within the ribbing of a cuff of a sleeve of a black sweater
belonging to Rickie Preddie. This stain also matched the DNA
of Damilola Taylor.
The discovery of the two bloodstains led to a prosecution
of the two brothers and they were eventually found guilty of
manslaughter in 2006. The Home Office review found that
human rather than systemic failure led to the omission of
examining relevant bloodstains in the first examination. The
reviewers identified a conflict between the pursuit of excel-
lence and the demand for urgent results.
This case shows that sometimes only minimal evidence is found after a
violent bloody attack. Although extensive blood loss may be obvious at
a scene where a victim has bled to death, it may not be discovered for
some time after the assaultlong after the offender has fled. Blood from
a stab wound may transfer very small quantities of blood to the cloth-
ing of an assailant. However, the age and quality of the stain will impact
the quality of a DNA profile. Old, degraded stains and those affected by
molds and fungi may yield little or no DNA.
3.2.3Semen and spermatozoa

Semen obtained from a healthy male individual has a quantity of sperma-
tozoa (sperm) that are rich in DNA. A specific extraction method is used
to separate the sperm from other cellular materials such as vaginal cells
obtained from medical swabs of the victim.
Differential extraction of seminal fluid with spermatozoa aims to
separate the seminal fraction (corresponding to the sperm of the male
donor) from the cellular fraction that may correspond to the female from
whom the swab was taken. Sometimes this extraction is incomplete or
unsuccessful and female cellular material is found in the seminal (sperm)
fraction and/or lysed spermatozoa are found in the cellular (nonsperm)
fraction. This crossover of material from two separate donors can compli-
cate interpretation.
Sperm is destroyed quickly in the relatively hostile environment of a
vagina. Many protocols recommend the taking of vaginal samples only if
the postcoital interval is less than 72 hours or three days (Mayntz-Press
et al., 2008). The literature notes that spermatozoa, although few in num-
ber, can sometimes persist in a vaginal canal longer than three days, but
the survival rates are longer in the cervix. It has occasionally been found
that spermatozoa survive more than a week in a deceased victim. The
ability to obtain a DNA profile of a semen donor from a vagina using
routine DNA profiling decreases rapidly after 24 to 36 hours following
coitus and it is usually not possible to perform a profile after 48 hours
(Saferstein, 2005).
Sperm loss after intercourse is due to vaginal drainage, menstrua-
tion, and the normal sperm degradation that occurs over time. Loss can
also occur during the multiple laboratory manipulations required for the
differential technique used to separate the sperm from the n onsperm
DNA fractions.
Case 2 illustrates the importance of the location of the semen in the
context of a case. The DNA profile of the semen that matches the accused
is not the only important factor.
Case 2
The prosecution alleged that the accused had forcible vaginal
sexual intercourse with the complainant. The accused stated that
he ejaculated on the underpants of the complainant and that the
act was consensual. Seminal stains were found on the front
of the complainants underpants, above the crotch line, and

yielded a DNA profile that matched the DNA of the accused.
The likelihood was estimated to be 1 in 10 billion that it was a
random match with DNA from another individual.
No semen or spermatozoa were detected in the medical
swabs from the vagina of the complainant. The location of the
semen stains did not correlate with the discharge of seminal
fluid from the vagina, which would be expected on the gus-
set or crotch area of the underwear. Moreover, the absence of
semen in the vaginal area did not accord with the proposition
of sexual intercourse with ejaculation within the specified
period of less than 24 hours. The accused was acquitted at trial.
Source: Case study from authors files.
Internal (vaginal, anal, or oral) swabs from a sexual assault complain-

ant are analyzed for spermatozoa using microscopic techniques. A medi-
cal officer often smears a medical swab onto a microscope slide so that the
slide can be examined for spermatozoa, leaving the swab intact for subse-
quent examination. This may improve efficiency for the forensic scientist
but creates potential for error and contamination because the evidence
now consists of two separate entities. Major English laboratories create
their own microscope slides from swabs to ensure a direct connection
between sperm detected and the swab and also retain a representative
sample for microscopy.
It should be noted that it is possible to have penile penetration into a
vagina without ejaculation of spermatozoa. A finger or other object may
also penetrate without deposition of detectable DNA material.
In all sexual offense cases, it is important that reference samples are
obtained from both victim and suspect(s) for exclusionary purposes.
Reference samples should also be obtained from any other male who may
have had sexual activity with the victim during the few days before the
alleged offense.
3.2.4Saliva
Nearly two decades ago, we learned that DNA can be extracted from
saliva deposited on postage stamps (Hopkins et al., 1994). Obtaining DNA
from saliva is now a routine process. The DNA obtained from saliva is not
present in the salivary excretions themselves. It is present in the mouth
(buccal) cells that are shed into the saliva. Thus, the DNA success rate on
saliva is variable because it is not possible to predict the quantity of mouth
cells in a saliva sample or stain.
On occasion, DNA may be recovered from drinking vessels, straws,

food, cigarette and cigar ends, saliva stains on gags, targeted areas on
headwear, envelope flaps, and licked stamps. The DNA from mouth cells
is prone to degradation due to the high numbers of bacteria in the mouth.
The DNA from saliva found in fizzy drink containers is particularly prone
to degradation due to the acidic nature of carbonated drinks that may
attack the DNA.
3.2.5Hair roots
Pulled hair samples that include the roots and/or cellular material from
the scalp and skin may be excellent sources of DNA. The roots of shed or
fallen hairs that are often found on clothing or at crime scenes contain
little cellular material. The hairs may provide mitochondrial DNA from
the shafts.
A number of successful solutions of cold cases were achieved by
profiling DNA obtained from hairs in cases where little other evidence
remained (see Chapter 7). Advances in technology allow examiners to
obtain nuclear DNA from naturally shed hair roots and even from hair
shafts. Thus hairs hold great evidentiary potential (Taupin, 2004).
3.2.6Dandruff and skin

Dandruff and the surface layers of skin may be suitable for nuclear DNA
analysis. DNA may be obtained from any nucleated epithelial cells found
in dandruff and the quantity has been estimated at 0.8 to 16.6 nanograms
(ng) of DNA per dandruff particle (Herber and Herold, 1998). Medical
conditions such as psoriasis and seborrhea dermatitis will also deposit
skin in a similar manner to dandruff. Dandruff remains a problem among
the normal healthy population.
Fingernail scrapings may be of value when someone scratches another
individual deeply enough to draw blood or remove skin.
3.2.7Nasal secretions
Used handkerchiefs can be valuable sources of relative large quantities of
cellular DNA from the nose area. Also, an area of clothing such as a sleeve
may been used by an accused or a victim to wipe his or her nose.
3.2.8Vaginal secretions
Vaginal fluid contains cells from the lining of the vagina. In a sexual
offense case, vaginal fluid may be found on the outside of a condom or on
an item used to sexually assault a victim. Vaginal cells may also be pres-
ent in a semen stain but cannot currently be differentiated from normal
epithelial (skin) cells.
3.2.9Sweat
Sweat is a liquid secretion that contains no cellular material. However,
certain areas of clothing such as the armpit of a shirt or the inner sole of
a shoe may contain a mixture of sweat and skin cells sloughed from the
body. It is believed that the sweat acts as a vector for the transfer of skin
cells onto clothing. Areas of clothing may hold a reservoir of DNA and
thus be valuable in identifying wearer DNA and determining the usual
wearer of a garment.
3.2.10Wearer DNA
Wearer DNA is deposited on clothing when it is worn. This DNA is depos-
ited through contact with the skin and consists of nucleated epithelial cells.
The usual wearer of a garment should be detected as the major source of
DNA on a garment, but minor DNA profiles of other individuals may also
be detected if the wearer had close contact or lent the garment to another
person. Nucleated cells from other body areas such as the eyes, nose, or
mouth also yield successful DNA profiles. The hands may transmit nucle-
ated cells to different parts of clothing.
A cold case from Australia (Case 3) that baffled police for nearly
12 years illustrates the potential evidentiary value of wearer DNA belong-
ing to an offender and the blood of a victim found on a pair of shoesthe
blood caused the offender to discard them (Hall, Kennedy, 2011).
Case 3
A heavily intoxicated 20-year-old man in Sydney, New South
Wales became involved in a fight with a man with a goatee
at a taxi stand in 1995. The victim was bashed and died from
severe injuries in a nearby car park. Three unidentified men
who came from a nearby nightclub were also involved in the
fracas. The victims wallet was stolen from his back pocket and
his running shoes were taken.
Two days later, a lawyer in an office a few blocks from the
murder scene looked out his office window and saw a pair of
running shoes on an awning. The lawyer gave the shoes to the
police and the DNA from the blood on the outside of the shoes
matched the DNA of the victim in the bashing.
The inside tongue of the shoes was sampled and another

DNA profile was obtained from this wearer DNA. The profile
did not match anyone on the database but it matched DNA
from blood found on the inside of the victims back pocket. The
police theorized the offender injured his knuckles in the alter-
cation. He discarded his shoes because they were covered with
blood and stole the victims shoes.
In 2008, Darren Paul Smith was arrested by Queensland
police for being drunk while riding a bicycle he had stolen after
drinking in a pub. The larceny offense in Queensland required
Smith to submit a sample for DNA profiling on the database.
The sample placed on the national database matched the DNA
profile from the wearer DNA on the tongue of the running
shoes and from the back pocket of the victims trousers. The
defense argued that the DNA in question came from transfer
after the relevant exhibits were put into the same exhibit bag.
The accused was found guilty by a jury and sentenced to a
minimum of 18 years in prison.
3.2.11Touch DNA
The first demonstration that simply touching an object will leave suf-
ficient amounts of DNA for a profile occurred in 1997 in a Melbourne
laboratory (van Oorschot and Jones, 1997). The research arose from an
unexpected DNA result in a criminal case. Since then, the amount of
literature about investigating and utilizing touch DNA as evidence has
risen exponentially.
This author was involved in the investigation of the rape of an elderly
woman in her own home (Case 4); touch DNA evidence led authorities to
the offender (Taupin, 2009).
Case 4
An unknown intruder armed with a knife raped an elderly
woman in her home in northern England. She was dragged by
the young male offender and subsequently found in the street
outside, dressed only in her bra and sweater. She could not
identify the offender beyond a general description. No semen
was found on the house carpets, medical swabs of the victim,
or the victims bloodstained sweater. Because she had been
dragged, there was potential for the deposits of touch DNA on

the relatively rough surfaces of the lace parts of the bra cups
and the areas were taped.
The DNA analysis revealed a major DNA profile cor-
responding to the female victim and a complete minor male
profile that was submitted to the national DNA database and
matched one convicted offender. He had recently been released
from prison and had just pawned three rings that belonged to
the victim. He pleaded guilty before trial.
It is recognized that some individuals may have more propensity to

shed DNA-containing cells than the rest of the population; these people are
called shedders (Lowe et al., 2002). Knowledge of an individuals shed-
ding characteristics may be useful in compiling general background data
in the interpretation of DNA trace evidence (Phipps and Petricevic, 2007).
Because we cannot determine when touch dells are deposited, the
degree of uncertainty about how the DNA may have been transferred
to the object and the relevance of findings may be difficult to assess.
In fact, the emphasis at trial may shift from whose DNA is this? to
how did this persons DNA get here? (Evett et al., 2002). The standard
considerations applicable to all types of trace evidence (transfer, persis-
tence, and recovery) are just as important for DNA evidence.
3.2.12Urine and feces

Urine and feces stains are not routinely examined because they do not usu-
ally contain enough cellular material sloughed from the lining of the ure-
thra or anal canal to obtain a DNA profile. Both body materials are waste
products. Only when sufficient cellular material may have been eliminated
by a person (blood from hemorrhoids, for example) should a routine DNA
analysis be considered. A scientist should be consulted and provided a
specific scenario to determine whether obtaining DNA is feasible.
3.2.13Emerging techniques
Innovative and novel analyses are mentioned routinely in newspapers
because they often capture the public imagination. Case 5 involved
the use of maggots to identify the body they consumed (de Lourdes
Chavez-Briones et al., 2012). As a result, insects at crime scenes may now
be investigated with more interest.
Case 5
In a wooded area, Mexican police found a body that was burned
beyond recognition and identification was not possible even by
DNA profiling. A woman had been abducted 10 weeks earlier
and her ring was found nearby. The face and neck of the body
were colonized by fly larva (maggots) that are frequently found
and collected from corpses, especially those found outdoors.
DNA typing was performed on the gastrointestinal con-
tents of the maggots and compared to DNA obtained from the
abducted womans father, with a probability of paternity of
99.685%. This was the first reported case of the use of human
DNA from maggots to identify a victim in a criminal matter.
Predicting externally visible characteristics such as hair and eye colors

(phenotyping) is an emerging field in DNA profiling. The HIrisplex test
that can predict both the eye and hair colors from DNA left at a scene was
reported recently (Walsh et al., 2013). This test is expected to be useful
when a perpetrator cannot be identified through DNA profiling.
3.3Reference samples
3.3.1Buccal scrapes
The inside of the cheek is scraped four or five times with one or more
plain sterile cotton or Dacron swabs to remove cells from the lining of
the mouth. This type of sample is considered less invasive than the tra
ditional reference blood sample and provides more than sufficient DNA
if taken properly.
3.3.2Blood
Blood samples are often collected from deceased or surviving motor vehi-
cle accident victims for medical and chemical tests and extra samples may
be taken for DNA profiling.
3.3.3Plucked hair samples

Plucked hairs are collected at mortuaries as part of the autopsy protocol
and can also be used for DNA profiling if the other body samples are
degraded or otherwise unsuitable.
3.3.4Personal belongings
Missing person cases rely on items regularly used by the missing per-
son such as toothbrushes and hair brushes. The DNA obtained from the
belongings is then compared with DNA from relatives. Care should be
taken in sampling belongings because they may have been used by some-
one other than the owner.
3.4Current profiling technique: Short

tandem repeats (STRs)
The genetic markers currently typed in most forensic biology laboratories
include autosomal short tandem repeats (STRs), mitochondrial DNA, and
Y chromosome STRs (Y-STRs). This section will discuss nuclear DNA pro-
filing using STRs (autosomal DNA profiling). The main steps in STR DNA
profiling are
Isolate the crime stain or other biological sample.

Separate the DNA and clean the sample from the other material.
Measure the quantity and quality of the DNA.
Target the specific areas of interest within the DNA molecule.
Produce multiple copies of the DNA pieces.
Sort the DNA pieces according to size.
Measure the sizes of the DNA pieces.
A major advantage of STR profiling is that many areas of a DNA molecule

can be examined simultaneously in systems called multiplexes, thus reduc-
ing the amount of time required for a result.
Forensic DNA profiling examines locations along a DNA molecule
that are highly variable from one individual to another and still have no
known functions. The original method of DNA profiling used by Alec
Jeffreys was called restriction fragment length polymorphism (RFLP) and
required relatively large amounts of DNA. The process was also complex
and time consuming (see Chapter1).
The currently used polymerase chain reaction (PCR) technique was
introduced in the 1990s and is useful for very small and degraded sam-
ples. PCR produces millions of copies of a particular area or sequence
of DNA and ensures the production of sufficient material for analysis.
It targets short repeated sequences or short tandem repeats (STRs) on a
chromosome that are variable with specific sequence primers. The STRs
are amplified many times so that they can be detected.
The DNA fragments that result from the process are then separated,
detected, and analyzed. A major advantage of this type of STR profiling
is that many areas of a DNA molecule can be examined simultaneously.
It should be noted, however, that the potential for error also increases
because of the many steps of the method and the human involvement at
each step.
The most common method of DNA profiling for forensic purposes
uses a variation of short unit repeat loci (STRs) on chromosomes in the
nucleus of the DNA molecule that are inherited from both parents. This
is called autosomal STR profiling. These profiles are known as genotypes,
as each STR is inherited from both the mother and the father. Humans
have 22 pairs of autosomal chromosomes that are not involved in deter-
mining sex. The remaining pair consists of the sex-determining X and
Y chromosomes.
The STRs (short unit repeat loci) in a DNA molecule are short unit
lengths of DNA that are repeated end to end. A reasonable number of
STR loci chosen (9 or more) can provide a high level of individualiza-
tion in the population chosen for the sample. STR markers have become
important tools for human identity testing and will continue to be used
for many years because of their high degree of variability, ease of use in
multiple amplifications, and implementation in national DNA databases.
A core set of STR loci allows national and international sharing of crimi-
nal DNA profiles.
STRs consist of regions of two to seven base pairs repeated in tandem.
Individual variations involve the number of repeats and/or the contents
of the repeats. A variation in the content of the repeats occurs as a change
in the base or as a deletion within a repeat unit. STRs used in forensics are
either tetra (four-time) or penta (five-time) repeats. Most forensic laborato-
ries use the four-base pair repeat systems. STRs are highly abundant and
well studied in the human genome, and their small size and the small size
range of the alleles facilitate typing from highly degraded, small quanti-
ties of starting material.
There are 9 core loci in the Australian DNA database system, 10 in
the U.K. system, and 13 CODIS core loci in the United States database.
Efforts are in progress to increase the number of loci examined in rou-
tine casework by using more sophisticated multiplexes such as the 16-loci
IdentifilerPlus (Wang et al., 2012).
The technique and the statistics used in autosomal STR testing are
well developed and national and regional DNA databases are in use in
many countries through statute. If no descriptive prefix precedes the
DNA profiling term, it can be accepted that the nuclear STR method of
DNA analysis described above has been used.
3.5Reading tables of alleles

A forensic DNA report may include a table of alleles that compares the
crime scene sample DNA profiles with reference sample profiles. These
tables most often also include samples from the accused and victim.
The alleles are designated by a forensic scientist at a particular locus and
correspond to the values detected at the locus for a specific profiling sys-
tem. The SGM Plus system examines 11 loci including the amelogenin
sex marker.
Table3.1 shows a table of alleles of three samples (one from the crime
scene, one reference sample from the accused, and a victim reference
sample). The crime scene sample also appears in the electropherograms
of Figure3.2 and Figure3.3. The crime scene sample has all the designated
alleles above the stochastic threshold (not low level, see Chapter5).
Across the top of the table are the names of the various loci examined.
The alleles detected by the test at each locus are identified by numbers
indicating short tandem repeats. An individual has two alleles at each
locus, one inherited from each parent. In some cases, however, only one
allele is detected. This is shown in Table 3.1 as 16,16 at locus vWA for
the victim, and is interpreted as inheritance of the same allele (16) from
each parent.
Amelogenin (amel) is one of the loci analyzed and is used to determine
the sex of the contributor. Males have X and Y chromosomes (X, Y in the
allele table). Females only have X chromosomes (X, X in the allele table).
An examination of the DNA profiles in Table 3.1 can determine
whether the accused or the victim could or could not be the source of stain
evidence. In this example, only the accused could have been the source.
The table depicts a simple example, as the crime stain profile appears to
involve a single source (no more than two alleles at each locus) from a
male contributor. Both the accused and the victim are males (X, Y at the
amelogenin locus).
The forensic report associated with the above table would typically
state that the victim was excluded from contributing to the crime scene
sample and that the accused matched it and cannot be excluded. The
report would also cite a statistic as to the probability of the match of
the accused DNA reference profile and the crime scene DNA profile.
The allele table is a summary of the results and does not show raw
data or interpretation details. In order to analyze raw data such as the elec-
tropherogram (profile), the case file notes must be reviewed. The extraction
and amplification dates, quantification values, and electropherograms of
the crime samples and reference samples should be retained in the case
file because the forensic DNA expert will have to explain this data to the
criminal justice professionals involved in the case.
48
Introduction to forensic DNA evidence for criminal justice professionals
Table3.1 Example of Allele Table Shown in Forensic DNA Report
Locus D3 vWA D16 D2 AMEL D8 D21 D18 D19 THO1 FGA
Crime stain 15,16 14,16 9,10 20,23 X,Y 12,15 28,31 12,15 14,15 7,9.3 24,26
Complainant reference 15,16 16,16 10,11 20,22 X,Y 11,12 29,29 12,15 14,14 7,9.3 24,25
Accused reference 15,16 14,16 9,10 20,23 X,Y 12,15 28,31 12,15 14,15 7,9.3 24,26
Notes: The names of each locus are along the top row in their usual abbreviated form. The alleles of the crime stain are represented
in the electropherograms of Figures3.2 and 3.3. The full names of each locus are shown in the figures. The alleles are the
numbers under each locus, except for the s ex-determining amelogenin marker (AMEL). The accused is included as a contribu-
tor to the crime stain; the complainant is excluded as a contributor.
Chapter three: DNA profiling basics
D3S1358 vWA D16S539 D2S1338
100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350
240
200
160
120
80
40
0
15 16 14 16 9 10 20 23
AMEL D8S1179 D21S11 D18S51

100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350
160
120
80
40
0
X Y 12 13 28 31 12 15
D19S433 TH01 FGA

100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350
200
160
120
80
40
0
14 15 7 9.3 24 26
Figure 3.2 Electropherogram (epg) of DNA profile. Three tones correspond to groups of loci recognized by fluorescent dyes. The
loci are designated across the top of each color. D3 is the first locus and vWA is the second locus. The alleles are numbered peaks
along the X (horizontal) axis, e.g., 15, 16 for D3. The X axis indicates the time the DNA fragments take to progress through the
capillary. The Y (vertical) axis is measured in relative fluorescent units and denotes the amount of DNA present.
49
50
D3S1358 vWA D16S539 D2S1338
100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350

3000
2000
1000
0
15 16 14 16 9 10 20 23
AMEL D8S1179 D21S11 D18S51

100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350
2400
1800
1200
600
0
X Y 12 13 28 31 12 15
D19S433 TH01 FGA

100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350
1600
1200
800
400
0
14 15 7 9.3 24 26
Figure 3.3 Electropherogram from Figure 3.2 with a larger scale on the Y axis so that the tops of the peaks can be observed.
Stutter peaks are present at all loci (but not designated) except THO1 and the amelogenin sex marker. The stutter peaks can be
viewed more clearly in Figure 3.2 due to the smaller Y axis scale.
The next section summarizes the basic steps in the DNA analysis of a
sample that form the basis for the table of alleles or the statistic provided
in a forensic report.
3.6Obtaining DNA profiles

The following steps are required for STR analysis of a biological sample
found on an exhibit:
Extraction of DNA from biological sample
Quantification of obtained DNA
Amplification of STR loci by polymerase chain reaction (PCR)
Separation and detection of amplification products by capillary
electrophoresis
3.6.1Controls
It is essential that negative and positive controls and reagent blanks are
processed through the analysis along with the sample in question to
ensure that the quality systems work. The processing of these controls is
required for every batch of DNA samples analyzed.
A reagent blank in a DNA profiling examination tests for the pos-
sible contamination of the reagents and/or supplies by an external DNA
source during sample preparation. If the reagent blank exhibits one or
more peaks above a certain threshold, it should be r e-amplified. If the typ-
ing results remain after re-amplification, all DNA samples associated with
the reagent blank should be considered inconclusive and re-extracted.
If this is not possible due to the consumption of the DNA, the situation
becomes a management issue. If the source of the contaminating DNA
does not appear to be in the samples, the contamination should be noted
in the report. If extraneous DNA is present in both the reagent blank and
the associated sample, the sample should be reported as inconclusive.
A positive control is used to determine the accuracy and consistency
of the amplification and capillary electrophoresis processes. The positive
control contains DNA from a known source with a known profile. If the
positive control does not exhibit the appropriate results, the samples asso-
ciated with the positive control should be considered inconclusive and
re-amplified.
The negative control (amplification blank) contains all the reagents
for the amplification mix but no DNA. The negative control tests for the
contamination of samples during the setup of the amplification reactions.
If the negative control exhibits unexplainable peaks above a certain thresh-
old that are not eliminated after re-injection, all samples associated with
the negative control should be considered inconclusive and re-amplified.
The negative control and reagent blank show whether contamination

was present in the reagents or introduced through the testing process.
The case-reporting scientist should analyze the control results to ensure
the data meet quality standards. A problem with controls should alert the
criminal justice professional and it may not be resolved unless an expert
witness peruses the raw scientific data.
3.6.2Extraction
The method of extraction of DNA from a sample depends on the nature of
the sample. Epithelial cells from touch DNA require a simpler and quicker
extraction to isolate the DNA; a more extensive method is required for
spermatozoa and hair roots.
Reference samples such as buccal swabs are simpler to extract than
crime scene samples and have r ecipe-based analysis protocols. Samples
from crime scenes require scientific judgment as to the method of extrac-
tion according to body origin. Samples may be found on material that
may inhibit a polymerase reaction (e.g., dyes on denim jeans) or be associ-
ated with a substance such as mold that may degrade the DNA. The DNA
profile from such materials may reflect inhibition or degradation of the
sample. Another extraction technique and a repeat analysis of the sample
may be required to obtain an optimum profile. Evidence of degradation or
inhibition of DNA in the sample may be observed in an electropherogram
(see Section 3.6.5).
The simplest steps in extraction begin with disrupting the cellu-
lar material to obtain the DNA and release other materials such as pro-
teins. All the n
on-DNA material is removed by adding detergents and
proteases. The DNA material is often obtained as a pellet after a centrifu-
gation (spinning down) process.
3.6.3Quantification
DNA purification methods cannot differentiate human DNA from other
DNA, for example, from bacteria and fungi. If a sample is not pristine, a
human-specific DNA quantification system must be used.
The purpose of the quantification step is to enable the addition of the
optimum amount of DNA to the reaction to achieve amplification. This step
uses a small part of the extracted DNA and compares it to a DNA standard
of known quantification. Ideally, all samples will have the same amount
of DNA added to the amplification mixture. Too much DNA will result
in o
ff-scale peaks, l ocus-to-locus peak imbalance, and split peaks. Too little
DNA will result in poor quality and low level profiles (see Chapter5).
If no DNA is quantified, some laboratories stop an examination at

this point. Other laboratories will proceed to the steps below in order to
attempt a DNA profile. The quantification step is known to be less sensi-
tive than the actual profiling step, so an attempt to achieve a result may
be made if the evidence is considered crucial to the case, even if no DNA
is quantified in this step. The absence of a quantifiable amount of DNA in
a sample should be denoted in the case file because it indicates that low
level DNA techniques may be required for further analysis and interpre-
tation (see Chapter5).
3.6.4Amplification
The two most important factors affecting amplification and success of
nuclear DNA testing are the DNA quantity and degradation or inhibition
of a sample. The amplification process is applicable only to the DNA of
humans or higher primates (probably not an issue in criminal cases).
If the amount of DNA extracted from forensic samples is too small
to be detected by standard profiling, the amount must be increased.
Amplification makes many copies of the DNA material at each locus. The
technique for the amplification of the samples after the DNA is extracted
and quantified is the polymerase chain reaction (PCR). It can be used
on very small and degraded samples. It also targets the STRs on a chro-
mosome that are variable with specific sequence primers. The STRs are
amplified many times so that they can be detected. The DNA fragments
that result from the process are then separated, detected, and analyzed.
The PCR amplification procedure has three steps:
Denaturing: DNA strands of the double helix are separated by heating.

Annealing: By reducing the temperature, short synthetic DNA primers
that flank the region to be amplified hybridize with the target DNA.
Extension: The temperature is changed again to an optimal level for
the polymerase and new DNA is synthesized by polymerization.
These steps constitute a cycle (generally 28 cycles but may be up to 34 for

low copy numbers) and allow production of more than a million copies of
the target DNA in a few hours.
STR primers are human-specific; if a profile is obtained, we can be
assured that it is human. However, if no profile is obtained or sequenced,
the DNA must be quantified to assure that it is human. The primers iden-
tify the relevant STR-DNA segments and then amplify (replicate) these
segments. Each primer is labeled with a fluorescent light-reactive dye to
allow laser detection in the next step.
3.6.5Separation and detection

The fragments of DNA produced in the steps above are separated by capil-
lary electrophoresis in a genetic analyzer. The fragments are forced by an
electrical field through a narrow capillary tube in which the larger frag-
ments move more slowly than smaller fragments. Under laser light, the
colored dyes produce fluorescent light that signals the presence of DNA.
A computer-operated camera detects the light as the fragments reach the
end of the capillary. Based on the color of the light and the time of travel
through the tube, a series of computer programs determines which alleles
are present at each locus.
Dedicated software is used to interpret the c omputer-generated data.
The intensity and position of each light emission, displayed as a peak on
an electropherogram (epg), is compared against standardized measures
of known size and amount that constitute a sample known as a ladder that
essentially serves as a reference. Peak heights are measured in relative
fluorescent units (RFUs).
The initial data produced by the fluorescent detection instrument is
processed through software such as Genescan. The raw data are then
compared to the sizing ladder and peaks (alleles) designated by software
such as Genotyper. Other analysts (generally two) independently con-
firm the presence of the alleles and note other issues such as technical
artifacts. The case-reporting scientist is responsible for reviewing the soft-
ware data and drawing conclusions about the data based on the quality of
the profile and the controls.
3.6.6Reading electropherograms
The electropherogram (for example, Figure3.2) can be viewed as a type
of graph with an X axis (horizontal scale) and a Y axis (vertical scale). The
positions of the peaks on the graph (distance to left or right) indicate how
long it took a specific allele to pass through the capillary tube, and this
indicates the length of the underlying DNA fragment. The numbers under
each peak are computer-generated labels that indicate which allele each
peak represents and how high each peak is relative to the baseline.
The smaller fragments are located toward the left side of the graph
and the fragment sizes increase in length toward the right side. This is
because it takes longer for larger fragments to migrate along the capillary
tube than it does for the smaller fragments. The X axis is measured in
time. The Y axis measures the intensity of the signal and thus the amount
of DNA. As noted above, the RFU is the unit of measurement of the peak
heights. It may be necessary to alter the printout scale if the peaks origi-
nally appear off scale on the Y axis. The heights of the peaks must be
determined (see Figure 3.3). Magnified versions such as Figure 3.2 are
useful to determine the morphologies of smaller peaks that may not be
observed readily in reduced versions such as Figure3.3.
Peaks representing alleles from the same person are expected to have
roughly the same heights throughout a sample. This is certainly true
for reference samples. However, in crime scene samples, degradation
and inhibition may alter the balance of peak heights (see Section 3.6.7).
Furthermore, mixtures of DNA from two or more contributors may also
alter the proper balance (see Chapter5).
Accredited forensic laboratories will have their own validated DNA
systems. The literature contains numerous peer-reviewed articles, books,
and reports on DNA analysis from extraction to electropherogram inter-
pretation that are suitable as references for forensic examiners (Buckleton
et al., 2004; Butler, 2005), although they are highly complex.
3.6.7Artifacts and other technical issues

Artifacts are peaks or other abnormalities in an electropherogram.
Technical artifacts are observed often and have been documented exten-
sively in the literature. Laboratories are required, as part of their quality
systems, to use protocols to distinguish artifacts from real DNA peaks.
An independent expert may be required for a second opinion if a profile
of interest exhibits many artifacts. The presence of numerous dye blobs,
spikes in the electropherogram, split peaks, and shoulders on peaks may
indicate a poor quality profile resulting from poor or sloppy analysis.
Artifacts such as stutter, however, can be expected. The most common
stutter peaks are four base pairs smaller than the primary peak or associ-
ated allele. They result from a slippage of the strand during the amplifica-
tion process, and are one repeat unit smaller than the designated allele
on the electropherogram. Occasionally a forward stutter peak (four base
pairs greater than its associated allele) will appear.
Stutter peaks are evaluated by examining the ratio of stutter peak
height to the height of the appropriate adjacent allele, expressed as a
percentage. The height of stutter peaks can vary by locus but should not
exceed 20% of each allele. Peaks in a greater stutter position may indicate
a mixture of contributors to a profile.
Figure3.4 shows examples of stutter peaks immediately before each
allele. Stutter peaks may also be observed in Figure3.2. They are less dis-
cernible due to the size of the Y axis in Figure3.3.
Large stutter peaks, especially forward types, may indicate that too
much DNA was analyzed for optimum results. Again, an expert should
be consulted to determine the quality of the profile and decide whether
the DNA extract should be re-analyzed.
56
D2S1338

290 300 310 320 330 340
3000
2000
1000
0
20 23
Figure 3.4 Stutter in an electropherogram. The D2 locus has two designated alleles: 20 and 23. Each allele has one stutter peak
four base pairs before it on the X axis that appear as a smaller peak.
Pull-up represents a failure of the analysis software to discriminate

the different fluorescent dye colors labeled to the primers. Pull-up can
be readily seen on an electropherogram as the peaks are in the same
X position as the designated allele from a different fluorescent dye. Again,
pull-up is most often observed when too much DNA is loaded onto a
capillary tube. Pull-up can be observed in Figure 3.2 and corresponds
to allele peaks at the X axis (locus D16) and at FGA. The pull-up in this
electropherogram is at a low level and is not an issue.
Spikes in an electropherogram are caused by fluctuation in voltage or
air bubbles in the capillary tube. They do not look like allele peaks and
should be readily visible. They are also usually observed in two colors.
Dye blobs are usually broader than real peaks and are thought to appear
when the fluorescent dye becomes detached from the DNA fragments.
An off-ladder allele may be marked OL on an electropherogram and
may represent an unusual variant that does not accord with the ladder
reference alleles. Alternatively, it could indicate a technical artifact. An
interesting case in which an o ff-ladder allele was presented to numerous
analysts who made different interpretations (Thompson, 2009) is pre-
sented in Chapter5.
Stutter and some other artifacts may complicate the interpretation of
mixture DNA profiles by masking real peaks. This is also discussed in
Chapter5.
Degradation of a DNA sample may often be observed on an electro-
pherogram. Little or no degradation occurs if crime scene samples are
well preserved and isolated from unfavorable conditions such as heat and
moisture. The longer fragments of DNA are more likely to be affected first,
and the consequence is that amplification may be partial or fail completely
compared to results with shorter fragments of DNA. This may produce a
ski slope effect in an electropherogram: the peaks toward the right side
of the diagram are noticeably smaller in height than the ones toward the
left-hand side. Sometimes the peak heights of larger fragments are too
low to be discerned from the baseline and thus only a partial profile can
be obtained.
Degraded samples are particularly problematic in mixture samples
because the two or more samples composing a mixture DNA profile may
have different levels of degradation and thus lead to different interpreta-
tions by analysts. Figure3.5 shows a degraded DNA profile; some of the
loci show no discernible alleles. OL indicates an off-ladder peak.
The shorter fragments in Figure 3.5 (toward the left side) amplified
better than the larger fragments to the right. The figure is a poor quality
profile showing low peak heights and d rop-out. In the authors opinion,
the result should not be reported without further analysis. The sample
may have been a t wo-person mixture due to the presence of three alleles
58
D3S1358 vWA D16S539 D2S1338
100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
240
180
120
60
0
14 18 16 17 18 12
AMEL D8S1179 D21S11 D18S51

100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
400
200
0
X Y 11 14 28 30 12 OL
D19S433 TH01 FGA

100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
400
300
200
100
0
14 15 21 22
Figure 3.5 Electropherogram showing degradation and inhibition of DNA by reducing sizes of fragments from left to right in the
diagram. An o ff-ladder allele is marked OL and a spike appears in the corresponding X position.
at locus vWA. The designated alleles at loci vWA, D16, D21, D18, and FGA
are all below 200 RFU. No peaks are present for loci D2S and THO1.
Inhibitors in the samples can affect the PCR amplification process.
Body fluids left on soil, sand, wood, or vegetable matter can c o-extract
with human DNA and prevent or affect amplification. Other substances
such as clothing dyes (denim jean dye is a notable example in this authors
experience) may contain polymerase inhibitors. Samples containing
inhibitors often produce electropherograms similar to one from degraded
DNA that shows the ski slope effect. It may not be possible to differentiate
the cause of the small or absent peaks on the right side of the graph. One
study explains the environmental and chemical degradation and PCR
inhibition on single source samples and mixtures (McCord et al., 2011).
There are alternatives to dealing with a compromised, poor quality
DNA profile. The original DNA extract can be cleaned up via several
methods (Butler, 2012). An extract colored with dye, for example, should
indicate to an analyst that inhibition of the DNA extract is a real possibility.
3.7Time required to obtain DNA profiles

Stating a specific amount of time to produce a profile is difficult because
it depends on the probative value of the evidence and importance and
complexity of the case. I mpact-based priority systems should be used to
maximize the evidential value and allow the most crucial evidence to be
examined first (Taupin and Cwiklik, 2010). A garment showing damage
and multiple stains of blood and semen may require bloodstain pattern
interpretation and damage analysis before testing for semen. Determining
the sampling methods and correct stains for DNA profiling can be per-
formed only after information is obtained from a physical examination of
the evidence.
The time for examination of the items and isolation of the pertinent
DNA may vary from about 30 minutes to a few hours. The recovery of
touch DNA is relatively simple and performed via tape lifting or swab-
bing an item. Recovery of semen from medical swabs or clothing is more
time consuming because the presence of spermatozoa must be confirmed
before analysis. This involves using a presumptive chemical screening
test on presumed semen stains on a garment followed by microscopy to
identify spermatozoa.
Table3.2 estimates the time required for each step of a DNA analy-
sis. These times are based on samples designated urgent. Crime scene
samples must always be analyzed with positive and negative controls to
ensure the quality system works correctly. That is why a single sample
cannot be analyzed in isolation to reduce the time waiting for results.
Table3.2 Analytical Times for DNA Stepsa

Step Time
Extraction 90 minutes for blood; overnight for semen and hairs
Quantification 3 hours
Amplification 3 hours
Separation and detection 2 hours
Analysis of fragments 2 hours for 2 scientists
Interpretation Variable, minimum of 1 hour
a Approximate as of 2012.
Samples are usually processed in batches that contain samples from

many cases. A batch of samples from multiple cases proceeds through the
extraction, amplification, and other steps. Checking during the steps is
vital to avoid transcription and other handling errors.
Today, many forensic laboratories worldwide have large DNA anal-
ysis backlogs. This is the major factor that prevents prompt receipt of
results in routine cases. Urgent cases, however, can be processed rela-
tively quickly if an impact-based priority system is followed. This author
was the reporting scientist in a case in which DNA profiles (including
wearer DNA) of three bloodstained items were obtained and interpreted
in 1.5 days (information from authors files). The case required overtime
work by the examining scientist and the DNA analysts. Some forensic
laboratories have backlogs of 6 to 12 months or more.
3.8Designating peaks
The peaks or alleles in an electropherogram or DNA profile are generally
first designated in a forensic laboratory by automated software. Computer
programs such as the GeneMapper and Genotyper are available to
licensed laboratories. The programs utilize an allelic ladder that is essen-
tially a sample that contains a sizing tool for reference against a crime
DNA sample.
After the peaks are designated by the computer program, the electro-
pherogram is again interpreted by one or more laboratory scientists. The
designated peaks are either affirmed or discarded according to the specific
laboratory guidelines. Most often this is the point when the case-reporting
scientist compares the crime DNA profile with reference profiles.
There are thresholds used in the interpretation of a DNA profile
and the designation of peaks. The analytical threshold is a level above
which a peak may be determined as real and distinguishable from noise.

Validation studies should be performed by the laboratory to determine
the analytical threshold in use. Sometimes a threshold is given a uniform
value of 50 or 100 RFU for every electropherogram. United Kingdom
laboratories generally determine the analytical threshold for each electro
pherogram based on the s ignal-to-noise ratio.
The stochastic threshold is another specification used in the inter-
pretation of an electropherogram, particularly with low level profiles
(discussed further in Chapter 5). Peaks below this threshold may exhibit
drop-out of one of the two alleles in a heterozygote.
After a DNA profile is obtained, the case-managing scientist deter-
mines the next step. Often the next step is a comparison of a crime scene
profile with one or more reference profiles. If the reference profile is
excluded from contributing to the crime scene profile, then no statistic
is generated. If the reference profile is included as contributing to the crime
scene profile, a statistic should be generated. Chapter4 discusses this topic.
Although DNA profiling is considered more objective than other
forensic studies, we can see that the discretion of a scientist still comes into
play when interpreting DNA profiles. Chapter8 discusses this subject in
more detail. Examiner discretion applies most particularly to mixtures or
low level DNA (see Chapter5). Single source DNA profiles with more than
adequate quantities should present little challenge. Chapter 5 also dis-
cusses an interesting case involving low level DNA presented to numer-
ous analysts who provided varying interpretations (Thompson, 2009).
3.9Case documentation and review

Laboratory accreditation generally requires a technical review and an
administrative review on every reported case. A technical review is
essentially a peer review performed by a scientist not involved in the case.
Independent conclusions are drawn from the data presented in the
case file. Any discrepancies should be resolved by an independent third
party (usually the manager of the unit).
The administrative review is generally performed by the manager of
the unit and ensures that all appropriate documentation is present in the
case file, quality systems have been observed, and the final report makes
sense and reflects the examinations performed. Figure 3.6 lists the con-
tents of a generalized case file relating to a DNA report from a crime sam-
ple. If any of the listed documentation is missing, the legal professional
should query its absence.
Forms for submitting exhibits to laboratory including case ID number

Chain of custody documentation for all exhibits from receipt to return
Requests for examinations by investigators
Details of reference DNA samples submitted for matter and reason for
submission (identification of suspects, victims, known and accepted parties who
may have deposited DNA, for example, a previous consensual partner in a
sexual offense case)
Communications such as emails to and from laboratory biologists, investigators,
and legal counsel
Scientific hypotheses formulated
Testing rationale including further testing if DNA results are not obtained for
certain samples
Time frame proposed including projected court dates and urgent investigative
deadlines
Examination notes relating to items of evidence (such as a pair of jeans)
including photographs and diagrams
Presumptive test data (for example, acid phosphatase for determining semen)
and results of searching for biological fluids
Confirmatory tests and results
Sampling details of exhibit for DNA analysis
Data sheets for each DNA sample including details of:
Extraction
Quantification (amount of DNA quantified in sample)
Amplification
Separation and detection
Accompanying controls
Electropherogram showing designated alleles and artifacts and thresholds used
Table of designated alleles for each sample
Determination of single source, mixture profile, partial profile, or low level for
each sample
Determination of suitability of crime scene profile for comparison (indicate poor
quality and/or insufficient alleles and possible need for reanalysis)
Comparison of crime scene samples with reference samples, bases for exclusion
or inclusion, and reasons
If inclusion, note population database used for further comparisons and reasons
Statistical calculations
Software calculations
Manual calculations used to check software results (may be required in
complex matters)
Control calculations (dummy data) to ensure software and manual
calculations work as expected
Technical review notes
Administrative review notes
Final copy of scientific report
Figure 3.6File documents that should be retained in cases involving forensic

DNA testing. They will be required for forensic reports.
References
Buckleton, J.S., Triggs, C.M., and Walsh, S.J., Eds. 2004. Forensic DNA Evidence
Interpretation. Boca Raton, FL: CRC Press.
Butler, J.M. 2005. Forensic DNA Typing: Biology, Technology, and Genetics of STR
Markers, 2nd ed. Burlington, MA: Elsevier Academic Press.
De Lourdes Chavez-Briones, M., Hernandez-Cortes, R., Diaz-Torres, P. et al. 2012.
Identification of human remains by DNA analysis of the gastrointestinal con-
tents of fly larvae. Journal of Forensic Sciences, online September 12.
Evett, I.W., Gill, P., Jackson, G.M. et al. 2002. Interpreting small quantities of DNA:
The hierarchy of propositions and the use of Bayesian networks. Journal of
Fleming, R.I. and Harbison, S. 2010. The development of a mRNA multiplex
RT-PCR assay for the definitive identification of body fluids. Forensic Science
Gill, P. and Buckleton, J.S. 2010. A universal strategy to interpret DNA profiles
Hall, L. 2011. Drunken cyclists DNA leads to cold case murder conviction. Sydney
Morning Herald, September 17.
Herber, B. and Herold, K. 1998. DNA typing of human dandruff. Journal of Forensic
Sciences, 43, 648656.
Hopkins, B., Williams, N.J., Webb, M.B.T. et al. 1994. The use of minisatellite vari-
ant repeat polymerase chain reaction (MV-PCR) to determine the source of
saliva on a used postage stamp. Journal of Forensic Sciences, 39, 526531.
Kennedy, L. 2011. If the shoe fits killer nabbed 16 years later. Sydney Morning
Herald, July 10.
Lowe, A., Murray, C., Whitaker, J. et al. 2002. The propensity of individuals to
deposit DNA and secondary transfer of low level DNA from individuals to
inert surfaces. Forensic Science International, 129, 2534.
Mayntz-Press, K.A., Sims, L.M., Hall, A. et al. 2008. Y-STR profiling in extended
interval ( 3 days) postcoital cervicovaginal samples. Journal of Forensic
Sciences, 53, 342348.
McCord, B., Opel, K., Funes, M. et al. 2011. An investigation of the effect of DNA
degradation and inhibition on PCR amplification of single source and mixed
samples. National Institute of Justice. http://nij.gov/publications/digest/
issue6.htm
Peel, C. and Gill, P. 2004. Attribution of DNA profiles to body fluid stains.
International Congress Series, 1261, 5355.
Petricevic, S.F., Bright, J.A., and Cockerton, S.L. 2006. DNA profiling of trace DNA
recovered from bedding. Forensic Science International, 159, 2126.
Rawley, A. and Caddy, B. 2007. Damilola Taylor: An Independent Review of Forensic
Examination of Evidence by the Forensic Science Service. London: United
Kingdom Home Office.
Saferstein, R., Ed. 2005. Forensic Science Handbook, 2nd ed., Vol. 11. New York:
Pearson Prentice Hall.
Taupin, J.M. 2004. Forensic hair morphology comparison: A dying art or junk
science? Science and Justice, 44, 95100.
Taupin, J.M. 2009. Targeting areas of clothing for touch DNA: Unrelated cloth-
ing damage analysis leads to the solution of an aggravated rape. European
Academy Forensic Sciences Meeting, Glasgow, September.
Taupin, J.M., and Cwiklik, C. 2010. Scientific Protocols for Forensic Examination of
Clothing. Boca Raton, FL: CRC Press.
Thompson, W.C. 2009. Painting the target around the matching profile: The Texas
sharpshooter fallacy in forensic DNA interpretation. Law Probability Risk, 8,
257276.
van Oorschot, R.A.H. and Jones, M.K. 1997. DNA fingerprints from fingerprints.
Nature, 387, 767.
Virkler, K. and Lednev, I.K. 2009. Analysis of body fluids for forensic purposes:
From laboratory testing to n on-destructive rapid confirmatory identification
at a crime scene. Forensic Science International, 188, 117.
Walsh, S., Liu, F., Wollstein, A. et al. 2013. The HIrisPlex system for simultaneous
prediction of hair and eye colour from DNA. Forensic Science International:
Genetics, 7, 98115.
Wang, D.Y., Chang, C.W., Lagace, R.E. et al. 2012. Developmental validation of the
AmpFlstr IdentifilerPlus PCR amplification kit: An established multiplex
assay with improved performance. Journal of Forensic Sciences, 57, 453465.
Watson, J.D. and Crick, F.H.C. 1953. A structure for deoxyribose nucleic acid.
Nature, 171, 737738.
Watson, J.D. and Crick, F.H.C. 1953. Genetical implications of the structure of
deoxyribose nucleic acid. Nature, 171, 964967.
Wickenheiser, R.A. 2002. Trace DNA: A review, discussion of theory, and applica-
tion of the transfer of trace quantities of DNA through skin contact. Journal of
chapter four
Evidential value and statistics
4.1Introduction
The ultimate power of DNA profiling is in its statistical discrimination.
It is not possible to perform a DNA profile of every person in a country,
so a statistical probability is determined by a scientist if a match is found.
This statistic is based on the frequency of each allele or STR at nine or
more areas (depending on the profiling system used in the jurisdiction)
and involves multiplication that gives rise to the high values often
observed. However, factors such as relatedness of the population and
sampling effects are most often used in the overall calculation.
Statistics in general, as mentioned in Chapter2, can cause difficulty
in interpretation for investigators, lawyers, and others in the criminal
justice system such as jury members. The advent of DNA profiling and
its inherent reliance on statistics led various organizations to recognize
this problem and produce guidelines for legal advocates.
A United States judicial body developed a manual on scientific evi-
dence (U.S. Federal Judicial Center, 2011). The manual was written with
the needs of a legal audience in mind and covers a range of related topics
including: data collection and presentation, comparisons, association and
causation, and DNA evidence.
The Royal Statistical Society in London also produced two practitioner
guides for lawyers litigating DNA evidence (Aitken et al., 2010; Puch-Solis
et al., 2012). Aitkins guide focuses on statistical analysis; P
uch-Solis et al.
cover DNA evidence.
The statistical concepts in DNA profiling evidence may be hard to
grasp, particularly for those with little mathematical background. The
authors experience finds that the concepts are difficult to communicate
without setting aside a considerable period of time for contemplation.
The criminal justice professional should try to understand the basic con-
cepts in DNA profiling but should also seek clarification regarding the
statistical meanings of the evidence in his or her case from the forensic
DNA expert.
The next section summarizes the main statistical issues in DNA pro-
filing evidence along with some pertinent case studies. More complex
cases involving mixtures from two or more unknown individuals, low
65
level DNA samples, and partial DNA profiles are addressed in Chapter5.
Y-STR profiling and mitochondrial DNA both require different statisti-
cal considerations. These are discussed in Chapters 6 and 7. Case 1 in
Section 4.3.3 briefly covers an interesting case.
4.2Interpreting DNA profiles

DNA profiling is a comparative technique. A laboratory compares the DNA
result of a crime sample with that obtained from a reference sample. If
the two profiles are different, the donor of the reference sample cannot
have shed the body fluid from which the crime sample was generated. If
the DNA profiles are the same, the result is a match. One of the presump-
tions in determining a match between a crime DNA profile and a refer-
ence DNA profile is that the peaks (or alleles) are designated correctly.
Three kinds of alleles appear in a crime stain profile (Gill et al., 2006):
1. Alleles that are unmistakable

2. Alleles that may be masked by artifacts
3. Alleles that have dropped out completely and cannot be detected
Points 2 and 3 are discussed in Chapters 5 and 8. The subjectivity of the ana-
lyst is certainly a factor in designating a peak (Dror and Hampikian, 2011).
Ensuring that appropriate biochemical and genetic tests are per-
formed will mean that the best result is obtained. It is better to analyze
further or replicate stains on an item to try to obtain a better result than to
perform statistical analyses on suboptimal results.
If two profiles match, the person who provided the sample or some-
one who also has the same DNA profile can be the source of the eviden-
tial material. The significance of the match is determined by a statistical
analysis. An in-depth discussion of the statistical approaches is beyond
the scope of this book but many detailed texts on this subject are avail-
able (Aitken et al., 2010; Balding, 2005; Evett and Weir, 1998; Puch-Solis
et al., 2012). The criminal justice professional is encouraged to be familiar
with the basic principles and some of the terminology. Appendix A is a
glossary of common terms.
4.3Statistical approaches and obtaining

final statistics
4.3.1Random match probability and likelihood ratio
The interpretation of DNA profiles when a match is found requires a
determination of the probability that a second copy of the DNA profile
Chapter four: Evidential value and statistics 67
will be present in a certain population. The forensic literature contains

much debate about how this probability should be derived.
The two methods in common use to report DNA profiles are the clas-
sical probability approach and the likelihood ratio approach (Buckleton,
2005). An appendix at the end of a laboratory report should provide infor-
mation about the derivations and meanings of the statistics applied to evi-
dence implicating the suspect. A laboratory analyzing evidence yielding
full single-source DNA profiles will use one of two statistics:
1. Random match probability based on genotype frequency estimates

2. Likelihood ratio based on the primary hypothesis that the suspect
is the source of the DNA profile versus the alternate hypothesis in
which an unrelated individual from the general population is the
DNA donor
Random match probability (RMP) is the chance of a random DNA profile

match within a given population and is the reciprocal of the DNA pro-
file frequency. A DNA profile frequency is estimated by determining the
genotype frequency for each locus and then multiplying the frequency
across all loci. Rare genotypes provide stronger evidence, and population
databases sorted by race will yield somewhat different results. However,
it is important to understand that the result is a representation of how rare
a DNA profile is in a population.
The profile probability approach presents the probability of the occur-
rence of an evidentiary DNA profile (E) under a stated hypothesis H0.
This hypothesis may be as simple as saying that the DNA profile is from
a person unrelated to the suspect. The probability is written formally as:
Pr (E/H0)
where Pr is the abbreviation for probability and the conditioning bar is an

abbreviation for given.
Under the approximation that profiles from unrelated people are
independent, this probability for a single stain is the frequency of occur-
rence of the profile in the population.
An extension of the profile probability approach works with the
probabilities of the evidence under two or more alternative hypotheses
about the source(s) of the profile. This is called the likelihood ratio (LR).
A typical analysis of a crime sample utilizes the prosecution hypothesis
Hp (the accused is the source of the DNA) and the defense hypothesis Hd
(the accused is not the source of the DNA).
The LR is becoming the preferred approach worldwide. If the LR is
greater than one, the evidence supports the first (prosecution) proposition;
if it is less than 1, it supports the second (defense) proposition. The LR can

cope with other factors such as the uncertainty about the number of con-
tributors in a mixed profile (see Chapter5 for mixture interpretation). An
overall composite likelihood ratio is obtained by multiplying the likeli-
hood ratios for each locus on the assumption that the defendants profile
matches the questioned profile for the prosecution hypothesis.
4.3.2Calculating frequencies
It is necessary to determine the genetic composition of the relevant popu-
lation with allowances for coancestry, sampling, and blood relative issues.
The frequency of genotypes among major populations in the relevant
location may have been determined and published in forensic journals or
may be calculated by a particular laboratory if it maintains a particular
ethnic data set and its results have been published and validated.
The size of a database for these calculations may be as small as 100
individuals and still be valid for making reliable projections about a geno-
types frequency in a larger population (Chakraborty, 1992). The adequacy
of the sampling allowance method and the allele counts should always be
assessed formally (Curran and Buckleton, 2011).
The product rule is the simplest statistical calculation regarding
DNA evidence and was developed from Mendels work (discussed in
Chapter1). If the particular population from which the DNA is postulated
is large enough, it is assumed that any random effects can be ignored.
The independence of the loci (DNA molecule areas from the profile), also
known as the HardyWeinberg equilibrium and linkage equilibrium,
is assumed.
Allelic frequencies from databases that are deemed to meet the
HardyWeinberg criteria of independence and random mating are used to
calculate the genotypic frequencies of each STR locus result. These geno-
typic frequencies are then multiplied together to generate an estimated
frequency of occurrence of the obtained DNA profile in the population to
which the database corresponds.
Likelihood ratio calculationsincorporate factors such as the inbreed-
ing coefficient of a particular ethnic database and sampling correction.
The calculation is based on the Bayes theorem and conditional probabil-
ity. The Bayesian approach is now the foremost alternative for forensic
disciplines, in the literature if not in actual practice.
Conditional probability can be stated simply: given that A occurs,
what is the probability that B occurs? Stated another way, the probability
of B is conditioned on the occurrence of event A.
The Bayesian approach is based on at least three ideas:
1. It is necessary to consider an alternative proposition in any evalua-

tion of a probability.
2. Scientific interpretation is based on the probability of the evidence
given the proposition.
3. The interpretation is also conditioned on the framework of
circumstances.
The Balding and Nichols sampling formula is a correction factor

incorporated into the statistics for the likelihood ratio. It accounts for the
fact that the frequency of the particular genotypes in a laboratory popula-
tion database came from a sample of the population, not the entire pop
ulation (sampling an entire population is not currently possible).
4.3.3Comparison of probability of exclusion and LR methods

The combined probability of inclusion (CPI) and the combined prob-
ability of exclusion (CPE) calculations are used by some laboratories to
indicate the statistical significance of results. CPI is the percentage of
the population that can be included in a profile; CPE is the percentage
of the population that can be excluded. The CPI and CPE calculations
are closely related: CPI is calculated by multiplying the probabilities of
inclusion from each locus, and CPE is calculated by subtracting the value
obtained from the CPI calculation from 1 (i.e., 1 CPI). This terminology
is most often observed in Y-STR typing and mitochondrial DNA reports
because likelihood ratios cannot be performed for these analyses (see
Chapters 6 and 7).
Case 1 (Ayturgrul v. The Queen, 2012) was an appeal to the High Court
in Australia after a DNA statistic was alleged to be inadmissible.
Case 1
The appellant was tried in the Supreme Court for murder and
was found guilty. The deceased and the appellant had been
in a relationship that ended more than two years before the
victim was stabbed to death. The prosecution case at trial was
circumstantial.
Mitochondrial DNA testing (see Chapter7) was performed
on a hair found on the deceaseds thumbnail. This test showed
that the accused could have been the donor of the hair and
two statistics were submitted. It was expected that 1 in 1,600
people in the general population would share the mitochon-
drial DNA profile or that 99.9% would be excluded. It was
alleged that the exclusion percentage was not permissible.
The Court of Criminal Appeal dismissed the appeal and

special leave was granted to appeal to the High Court that then
held that the appellant did not demonstrate that the probative
value was outweighed by the danger of unfair prejudice.
It should be noted that mitochondrial DNA typing and Y -STR pro-

filing use different techniques from autosomal STR DNA profiling, and
the derivation of the statistical significance is different. The techniques
are less discriminatory than autosomal STR DNA profiling due to the
method of inheritance of haplotypeseither from the maternal line
(mitochondrial) or from the paternal line (Y-STR). The considerations
of haplotype frequencies and the way they are reported necessitate the
counting approach (see Chapters 6 and 7). The strength of the evidence
depends on the sizes of the databases.
The probability of exclusion, or random man not excluded (RMNE), or
the complementary probability of inclusion entails a binary view of alleles,
meaning that alleles are only present or absent. Furthermore, if they are
present, they are observed. If alleles are found where there is a possibility
of stochastic effects, laboratories may omit the inconvenient loci from their
calculations (Gill et al., 2006). Such a calculation incorrectly implies that
among the random men considered for comparison, only the same loci
as those considered for the suspect in question would be used for inclu-
sion or exclusion (see Chapter5 for low level DNA techniques).
Two methods of statistical significance were presented in the O.J.
Simpson case in California (Weir, 1995). The prosecution wished to use
the LR and the defense wanted to use the RMNE. The final result was
that the court heard both methods and ruled that the LR method was pref-
erable. Also see Chapter1 for a discussion of this case.
Clayton and Buckleton (2005) summarized the advantages and disad-
vantages of each approach. Full discussions of the various methods of inter-
preting evidence can be found in comprehensive texts (Buckleton, 2005;
Balding, 2005). According to the DNA Commission of the International
Society of Forensic Genetics (Gill et al., 2006), the scientific community has
a responsibility to support improvement of standards of scientific reason-
ing in the courtroom. This implies that concepts such as likelihood ratios,
whether difficult to convey or not, are the methods of choice for the statis-
tical evaluation of DNA profiles.
Computer software is available to forensic laboratories for calculating
statistics such as likelihood ratios. Some laboratories may perform manual
calculations to check their results, although the calculations may be very
demanding. Each particular laboratory must have validated the popula-
tion databases and genotype frequencies it uses in forensic calculations.
4.3.4Identity and rarity

It is important to note that statistical analyses can never lead to absolute
conclusions. DNA evidence is essentially probabilistic as shown above
and an expert witness should never denote an individual as the donor of a
genetic material from which DNA was produced. There is a growing real-
ization that all forensic science evidence is probabilistic and no current
forensic technology supports the unique identification of an individual.
Other forensic science disciplines follow binary match or no-match sys-
tems and this transparency deficit is being addressed (National Research
Council, 2009; Fingerprint Inquiry, 2011).
Two authors (Saks and Koehler, 2005) described the genetics-based
model of DNA profiling as highlighting the deficiencies in other forensic
disciplines in which untested assumptions and s emi-informed guesswork
are replaced by a sound scientific framework and justifiable protocols.
The statistics quoted in forensic reports for DNA profiles are often rarer
than 1 in 1 trillion, a number that is greater than the population in the
world (currently 6 billion). These statistics appear incredulous to many peo-
ple and their method of derivation difficult to understand. It is hoped that
this text explains that the statistics in most criminal cases are derived accord-
ing to assumptions made both in the comparison of DNA profiles, and the
quality of the profile itself (complete or partial/low level/mixture). It is also
the probability of the DNA profile occurring in a particular population,
not the probability of the case hypothesis (see Section 4.4 for legal fallacies).
An interesting example of how statistics can be readily misinterpreted
is the famous (at least in statistical circles) birthday problem. This partic-
ular problem has been used to illustrate misconceptions in DNA database
matches (Weir, 2007; Kaye, 2009). Assume that equal numbers of people are
born every day of the year. Then the random match probability for a par-
ticular birthday is 1/365. However, there is over a 50% probability that two
people in a group of 23 or more share a birthday. How could this be? This
is because there are 253 pairs of people in a group of 23 and the particu-
lar birthday is not specified. When translated to DNA issues, the birthday
problem has to do with multiple occurrences of any profile, not one par-
ticular profile (Weir, 2007).
4.4Legal fallacies
Using unfamiliar terminology plus difficulties in statistical interpreta-
tion may lead a legal professional to translate results to a wider perspec-
tive that may not be valid. Two well-known fallacies are common in the
legal community and sometimes even in the news media. The prosecutors
fallacy is also called the fallacy of the transposed conditional. This fallacy
translates the chance probability of a crime stain match to the probability

of innocence. For example, say there is a 1 in 100,000 chance probability of
a match in a city of 1 million people. The prosecution fallacy is to say there
is a probability of innocence of 1 in 100,000. The defense fallacy in this par-
ticular situation is to say the probability of guilt is 1 in 10.
Suppose a crime is committed in London (population about 7 million)
and a crime scene profile has a likelihood ratio (LR) of 1 in 1 million. The
prosecutor might say that the odds are a million to one in favor of the
defendant being guilty. However, based on population size, about seven
people in the city are expected to match the profile so it can be argued that
the odds are actually 7 to 1 in favor of innocence. The defense fallacy unre-
alistically assumes that each of the 7 people has equal probability of guilt.
An often-quoted case from England (R. v Deen, 1994; P uch-Solis et al.,
2012) illustrates the prosecutors fallacy. Deen was an early DNA case in
which the random match probability was quoted as 1 in 3 million.
Prosecutor: So the likelihood of this being any other man but Andrew
Deen is 1 in 3 million?
Expert: In 3 million, yes.
Prosecutor: You are a scientist doing this research. At the end of this
appeal a jury are going to be asked whether they are sure that it is
Andrew Deen who committed this particular rape in relation to
Miss W. On the figure which you have established according to your
research, the possibility of it being anybody else being 1 in 3 million,
what is your conclusion?
Expert: My conclusion is that the semen originated from Andrew Deen.
Prosecutor: Are you sure of that?
Expert: Yes.
The basic fallacy is contained in the first question when the attorney asks
the probability of the accused being the source of the DNA profile; the
attorney should have asked about the probability of the evidence. It is
the jurys responsibility to decide whether factual propositions have been
established by the evidence, not the expert.
Having been asked the wrong question, the expert in Deen con-
founded the fallacy, even to the extent of pronouncing himself sure that
Deen was the source of the stain. In fact, a random match probability of
1 in 3 million implies that about 20 people in the UK would be expected
to share the same profile.
The prosecution fallacy (transposing the conditional) may be described
by two simple statements (Aitken et al., 2010):
1. If I am a monkey, I have two arms and legs.

2. If I have two arms and legs, I am a monkey.
This logic problem can be avoided by using the LR strictly as quoted

in the forensic report. The probability of the evidence based on the
hypothesis should not be translated to the probability of the hypothesis
itself. It is also helpful to remember that DNA profiling evidence provides
only the probability of a match of DNA profiles in the relevant population,
not the probability that a particular person committed the crime. As will
be repeated throughout this book, DNA is only one piece of evidence in
a crime.
Limitations of the evidence must be described. The question of how
the DNA was transferred is one for the jury to consider. The scientists
main role is to outline the various modes of transfer that exist and advise
on the relative risks associated with the modes (Gill and Buckleton, 2010).
The uncertainties about the mode of transfer increase with touch DNA
evidenceevidence that cannot be associated with a particular body fluid
(Buckleton, 2009).
4.5Understanding reports: Common phrases

and their meanings
Identifying the strengths and limitations of facts and opinions is a corner-
stone of forensic science. Any forensic report or testimony should convey
the limitations of all tests and all the evidence. All conclusions, assump-
tions made, and inferences should be enunciated and clearly explained.
Differences or similarities between evidence and reference samples
should be explained as actual differences or similarities inherent in the
evidence or as consequences caused by imprecision of the test system
limitations. All alternative explanations (such as different hypotheses
proposed) should also be conveyed in the report or testimony.
4.5.1Inclusion and exclusion

Scientific statements should clearly support or refute a finding or state
that the result is not possible due to the limitations of the hypotheses pro-
posed. Case 2 from Western Australia (Merritt, 2010) shows how miscon-
ceptions may arise from the wording of forensic statements.
Case 2
ixteen-year-old Patrick Waring was accused of rape, spent a
S
year in detention, and was exonerated in 2007. The forensic
report stated that the accused could not be excluded from
the DNA profile taken from the victims underwear.
Not only is this poor English (double negative) but the

scientific use of excluded in a forensic laboratory context
is unclear to a lay person. A defense expert re-examined the
mixed DNA profile and concluded that no DNA evidence
linked the accused with the victim.
It can be confusing to the legal practitioner to delineate the excluded,

inconclusive, and not excluded terms when the rationale for assigning a spe-
cific term to a specific DNA profile is not explained in the report. The
terminology is especially problematic when the possible contributing
DNA cannot be accorded a statistical weighting, either when it is denoted
as inconclusive or not excluded. This implies the evidential value is similar
whether a result is inconclusive or not excluded.
One strength of DNA evidence is its high discrimination power and
thus power to exclude; another is the ability to provide statistics for the
probability of inclusion. When one or both of these strengths are absent,
the DNA evidence becomes commensurately limited.
An illuminating study on an adjudicated criminal case (Dror and
Hampikian, 2011; Geddes, 2010) involved a DNA mixture from a gang
rape. Two analysts from the original laboratory stated that Suspect 3s
DNA profile could not be excluded from the mixture profile. Both pro-
files were presented to 17 DNA analysts from the same accredited govern-
ment laboratory without contextual information.
The results of the analysts were not consistent. Only one analyst
stated that Suspect 3 cannot be excluded. Of the remaining 16, 4 stated
inconclusive and 12 stated excluded. The authors of the study sug-
gested subjectivity was present in mixture interpretation. They also note
that bias may have originally occurred since the results were not con-
sistent with the original result obtained by analysts who had contextual
information. See Chapter8 for more discussion on this case.
If any possible individual contributing DNA to a mixed sample is
deemed an inclusion, an associated statistical analysis must support that
inclusion. According to the Scientific Working Group on DNA Analysis
Methods (SWGDAM) Interpretation Guideline 4.1, a laboratory must per-
form statistical analysis in support of any inclusion in the context of a
case, irrespective of the number of alleles detected and the quantitative
value of the statistical analysis (Buckleton et al., 2007; SWGDAM, 2010).
Inclusion, included, and cannot be excluded all convey the same meaning
and thus statistical frequencies must be reported with any statement that
includes them. According to Section 3.6 of the SWGDAM interpretation
guidelines, if the known individual cannot be excluded from the profile,
he or she must be included (SWGDAM, 2010). Furthermore, Section 4.3
states that the laboratory must not use inconclusive/unreportable data

(e.g., at individual loci or an entire multi-locus profile) in statistical analy-
sis. Guideline 4.1 states specifically that the laboratory must perform
statistical analysis in support of any inclusion that is determined to be
relevant in the context of the case, irrespective of the number of alleles
detected and the quantitative value of the statistical analysis.
4.5.2Declared contributor
If an individual is accepted by all parties as a contributor to the DNA
detected, he or she is a declared contributor. Often in sexual offense cases in
which an intimate medical swab from a complainant reveals female and
male DNA, the complainant is considered a declared contributor to the
mixed DNA present.
It is important in a criminal case to obtain as many reference DNA
samples as required for elimination purposes and for determination of
declared contributors. The principles of mixture analysis should be borne
in mind for semen stains from clothing and intimate medical samples from
complainants in rape cases as shown in Case 3 (Thompson et al., 2003).
Case 3
An 11-year-old girl was raped by the pool of her home in
Oklahoma in 1991. Timothy Durham was a local resident with
a record of criminal violations and the police focused on him.
The victim identification and hair comparison evidence was
inconclusive but a semen stain on the victims swimsuit alleg-
edly matched the DNA (DQ-alpha) of Durham.
Despite 11 alibi witnesses who said he was in a different
state at the time of the crime, Durham was convicted in 1993
and sentenced to over 3,000 years in jail. In 1996, he contacted
the Innocence Project and asked for further DNA testing of the
semen stain. The new tests showed that Durham did not share
the DQ-alpha type present in the semen stain and he was also
excluded at several other genetic loci. The initial DQ-alpha test
was shown to be a false positive because the laboratory failed
to separate completely the male and female donor samples dur-
ing the differential extraction of the semen stain (Section 3.2.3
in Chapter3 covers differential extraction).
The victims alleles when combined with those of the true
rapist matched the alleles of Durham. The laboratory mistook
this mixture for a single source. Durham was released in 1997.
When calculating LRs for a two-person mixture, the declared con-

tributor is included in both the defense and prosecution propositions. If a
declared contributor can be assumed to be part of a mixture, the issue can
be resolved into a single additional contributor.
4.5.3Verbal descriptors
The verbal descriptor scale was devised by Ian Evett and Bruce Weir
in England in the late 1990s and later expanded (Evett and Weir, 1998;
Buckleton et al., 2005). The scale is commonly used in England and
Australia. It is designed to provide an indication of the value of a particu-
lar LR, using terminology intended to be consistent between scientists.
The scale ranges from extremely strong for the prosecution proposition (LR
greater than 1 million) to inconclusive (LR of 1) to extremely strong support
for the defense proposition if the LR is less than .000001.
Two advantages of DNA profiling (not found with other forensic
disciplines) are its high discrimination power and the generation of statistics.
This author believes that after a verbal descriptor is applied, the descrip-
tor has a chance of being preferred as a simpler option by a n onscientific
reader. This may then reduce what is a comparative analysis of two propo-
sitionsinherent to the LR and thus evidential valueto an unwarranted
conclusion. The situation is particularly confusing when mixture DNA
profiles must be considered.
4.6Sampling correction and uncertainty

A sampling (or size bias) correction is used to allow for the limited data-
base sizes used in frequency calculations. Various methods are used in
different jurisdictions. A common method uses the Balding and Nichols
equation that adds the case profiles as additional empirical data (Balding
and Nichols, 1994).
When some laboratories calculate LRs, a sampling uncertainty is also
factored into the result. The LR estimates are calculated from a sample of
the population. Data from another sample of the same population may
yield a different estimate. The method allows determination of the best
estimate and a lower value of LR that has a 99% probability of being lower
than the true value. This means that there is a 1% probability that the
true LR is greater than the value quoted. The true value of the LR can
be determined only if sampling uncertainty can be eliminated. Finding
a true value would require DNA typing of an entire population which is
currently not possible.
4.7Relevant population and impact

on statistical value
When determining LRs, two alternative propositions must be considered:
(1) the defense proposition is that the person of interest is not the donor
and (2) someone else contributed the DNA. In many cases the someone
else is of an unknown ethnic or racial origin. However, the defense may
wish to state (or the parties may agree to) designation of a particular
ethnic source of the suspect, e.g., Victorian Caucasian.
The most populous databases in Western countries such as the United
Kingdom, the United States, and Australia are Caucasian. Calculations
for other ethnic groups can be performed if the defense or other party
desires. Some ethnic groups may still be isolated genetically and cultur-
ally or essentially incorporated into the general population due to current
society norms. Consequently it is important to ensure that the statistical
calculations are performed on the relevant population.
An example of new ideas surrounding population databases is the
segregation of the Australian aboriginal population data along contem-
porary state and territory lines. This appears to mask the diversity that
exists within this subpopulation. Datasets collected among more tradi-
tional lines may be more appropriate, particularly to distinguish the most
genetically differentiated populations residing in the north of the conti-
nent (Goetz et al., 2008; Walsh et al., 2007).
The co-ancestry coefficient addresses the fact that two people within
an ethnic group are more likely to have similar genotypes than two people
from different ethnic groups. This is represented as Fst or theta.
The use of general population frequencies and the product rule will
disadvantage a suspect because the general population may not necessar-
ily display the same frequencies as the subpopulation (Buckleton, 2005).
4.8Relatives
Further analytical work must be performed if the suggestion is made that
a blood relative (e.g., a brother) of the accused is the true perpetrator. This
is the so-called brother defense. If it is possible to obtain a reference sam-
ple from the relative and full DNA profiles have been developed, they can
be compared and thus reduce the need for further statistical work. If a
reference sample is not obtainable, calculations based on the relatedness
are required. Mendels theory of heredity (see Chapter1) proposes that a
parent is equally likely to pass on either of their two alleles to offspring.
The inbreeding coefficient Fst is 0.25 for siblings and 0.0625 for cousins,
and the recognized reference (Balding and Nichols, 1994) has the values
for the most common relationships. Many forensic laboratories have pack-
ages that calculate the statistics for relatives but they are applicable only
to single source profiles. SWGDAM Guideline 5.2.3 (2010) covers relatives.
References
Aitken, C., Roberts, P. and Jackson, G. 2010. Practitioner Guide 1: Fundamentals
of Probability and Statistical Evidence in Criminal Proceedings. London: Royal
Statistical Society (available online).
Ayturgrul v. The Queen. High Court of Australia, HCA 15, April 18, 2012.
Balding, D.J. 2005. Weight of Evidence for Forensic DNA Profiles. Chichester: John
Wiley & Sons.
Balding, D.J. and Nichols, R.A. 1994. DNA profile match probability calculation:
How to allow for population stratification, relatedness, database selection,
and single bands. Forensic Science International, 64, 125, 1994.
Buckleton, J. 2005. A framework for interpreting evidence. In Buckleton, J. et al.,
Eds., Forensic DNA Evidence Interpretation. Boca Raton, FL: CRC Press, 2763.
Buckleton, J. 2009. Validation issues around typing of low level DNA. Forensic
Science International: Genetics, 3, 225260.
Buckleton, J., Curran, J., and Gill, P. 2007. Toward understanding the effects of
uncertainty in the number of contributors to DNA stains. Forensic Science
Buckleton, J., Triggs, C.M., and Walsh, C.J., Eds. 2005. Forensic DNA Evidence
Interpretation. Boca Raton, FL: CRC Press.
Chakraborty, R. 1992. Sample size requirements for addressing the population
genetic issues of forensic use of DNA typing. Human Biology, 64, 141159.
Clayton, T. and Buckleton, J. 2005. Mixtures. In Buckleton, J. et al., Eds., Forensic
DNA Evidence Interpretation. Boca Raton, FL: CRC Press.
Curran, J. and Buckleton, J. 2011. An investigation into the performance of meth-
ods for adjusting for sampling uncertainty in DNA likelihood ratio calcula-
tions. Forensic Science International: Genetics, 5, 512.
Dror, I.E. and Hampikian, G. 2011. Subjectivity and bias in forensic DNA mixture
interpretation. Science and Justice, 51, 204208.
Evett, I.W. and Weir, B.S. 1998. Interpreting DNA Evidence, Statistical Genetics for
Forensic Scientists. Sunderland, MA: Sinauer.
Fingerprint Inquiry. 2011. Inquiry Report. Edinburgh. http://www.
thefingerprintinquiryscotland.org.uk/inquiry/3041.html
Geddes, L. 2010. Fallible DNA evidence can mean prison or freedom. New Scientist,
August. http://www.newscientist.com/article/mg20727733.500
Gill, P. 2001. Low copy number DNA profiling. Croatian Medical Journal, 42, 229232.
Gill, P. 2002. Role of short tandem repeat DNA in forensic casework in the UK: Past,
present and future perspectives. BioTechniques, 32, 366385.
International: Genetics., 4, 221227.
Gill, P., Brenner, C.H., Buckleton, J.S. et al. 2006. DNA Commission of the
International Society of Forensic Genetics: Recommendations on the inter-
pretation of mixtures. Forensic Science International, 160, 90101.
Gill, P., Whitaker, J., Flaxman, C. et al. 2000. An investigation of the rigor of inter-
pretation rules for STRs derived from less than 100 pg of DNA. Forensic
Science International, 112, 1740.
Goetz, R., West, J., Walsh, S.J. et al. 2008. Population data from the NSW Aboriginal
Australian sub-population for the Profiler Plus autosomal short tandem
repeat (STR) loci. Forensic Science International, 175, 235237.
Kaye, D. 2009. Trawling DNA databases for partial matches: What is the FBI afraid
of? Cornell Journal of Law and Public Policy, 19, 145171.
Merritt, C. 2010. Film shines a light on the dangers of DNA dependence. The
Australian, May 21.
National Research Council. 2006. The Evaluation of Forensic DNA Evidence.
Washington: National Academies Press, p. 130.
National Research Council. 2009. Strengthening Forensic Science in the United States:
A Path Forward. Washington: National Academies Press.
Puch-Solis, R., Roberts, P., Pope, S. et al. 2012. Practitioner Guide 2: Assessing the
Probative Value of DNA Evidence. London: Royal Statistical Society (available
online).
R. v Deen, Court of Appeals, U.K. The Times, January 10, 1994.
Saks, M.J. and Koehler, J.J. 2005. The coming paradigm shift in forensic identifica-
tion science. Science, 309, 892.
SWGDAM (Scientific Working Group on DNA Analysis Methods). 2010. Interpretation
Guidelines for Autosomal STR Typing by Forensic DNA Testing Laboratories. FBI
website: http://www.fbi.gov/about-us/lab/codcdis/swgdam.pdf
Thompson, W.C. 2003. Review of DNA evidence in State of Texas v. Josiah Sutton.
February 6. http://www.scientific.org/archive/Thompson%20Report.pdf
U.S. Federal Judicial Center. 2011. Reference Manual on Scientific Evidence, 3rd ed.
Washington: National Academies Press.
Walsh S.J., Mitchell, R.J., Curran, J.M. et al. 2006. The extent of substructure in
the indigenous Australian aboriginal population and its impact on DNA evi-
dence interpretation. International Congress Series, 1288, 382384
Weir, B.S. 1995. DNA statistics in the Simpson matter. Nature Genetics, 11, 365368.
Weir, B. 2007. The rarity of DNA profiles. Annals Applied Statistics, 1(2), 358-370.
chapter five
Partial profiles, low levels,

and mixtures
DNA profiling is often considered by the lay public to generate complete
profiles, but crime scene profiles are not produced by pristine reference
samples. The result may be partial and degraded, at low level, or consist
of mixtures from multiple contributors.
Special considerations exist when a DNA profile is not complete or
comes from more than one source. The legal practitioner should be alert
to the issues explained below. A lower-than-expected statistical value that
implicates a client (for example, less than a default value typically in the
billions) should always be investigated further. Report notations indicat-
ing a partial profile, mixture, low level, or trace amount should also be
explored further.
5.1Partial profiles
A full DNA profile is always the aim of profiling analysis. However, an
incomplete or partial DNA profile may be obtained due to degradation
of the DNA in the sample, insufficient quantity or quality of the sample,
or a combination of these factors. A statistical evaluation involving a par-
tial DNA profile will generally have less statistical value than a complete
profile. The Crown Prosecution Service in England reported that half the
DNA profiles yielded from samples recovered from crime scenes are par-
tial profiles (2010).
A partial DNA profile may indicate low levels of DNA. When a sam-
ple contains small amounts of DNA, the larger fragments of DNA may
fail to amplify; only the smaller fragments amplify and the results at the
higher end of the electropherogram are missing.
A partial DNA profile may also indicate degradation or inhibition of
the DNA. Degradation of a DNA molecule occurs over time, particularly
when subjected to heat, sunlight, water, and/or bacteria (see Chapter3).
Thus it is not uncommon to see partial DNA profiles in cold cases or from
outdoor crime scenes.
Degradation is often signaled by the ski slope effect whereby
the heights of the alleles decline toward the right of the graph in the
81
electropherogram, like going from the top to the bottom on a mountain or

ski slope. Sometimes the alleles toward the right side of the graph disap-
pear. This can also be the situation with inhibited DNA (Chapter3). The
alleles toward the left side of the graph (smaller DNA fragments) may still
generate good height without being considered low level. Thus it may be
prudent to have a DNA analyst explain a particular profile in detail.
Figure 3.5 (Chapter 3) and Figure 5.1 are both partial DNA profiles
exhibiting degradation or inhibition of DNA molecules.
5.2Low level and suboptimal profiles

There has been increasing debate in the forensic science and legal com-
munities regarding the analysis, interpretation, and meaning in cases of
low amounts of DNA. Forensic DNA is a complex field of science and
the debates have not been resolved. Every report and laboratory should
acknowledge the complex meanings of trace or low level DNA, define the
parameters in which the results may be interpreted with confidence, and
communicate these limitations to the criminal judicial system.
5.2.1Definitions
There is a proposal that there should not be a definitive line between what
is considered low level DNA and conventional autosomal STR typing (Gill
and Buckleton, 2010). The literature cites several definitions:
Amount of DNA tested in the PCR reaction (for example, less than
200 picograms) based on assay quantification
Increasing the number of PCR cycles beyond 28
DNA profile appearance that exhibits stochastic effects (see below)
Low level DNA is now generally accepted as applying to situations in

which the amount of DNA available is limited or interpreting the result-
ing profile may require more considerations than profiles generated using
higher amounts of DNA. Low level DNA may also be defined as any sam-
ple that falls below recommended thresholds at any stage of the analysis,
from sample detection through to profile interpretation, and cannot be
defined by a precise picogram (extremely small weight, abbreviated as pg)
level (van Oorshot et al., 2010).
These profiles are considered suboptimal and should be designated
as such in any report or testimony. Data reliability is inferior and thus
additional measures must be taken to accurately reflect the sample exam-
ined (Butler, 2012).
Until the past decade or so, DNA was recovered only from visible
stains such as blood splatters and semen stains. Now technology has
Chapter five: Partial profiles, low levels, and mixtures
D3S1358 vWA D16S539 D2S1338
100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
160
120
80
40
0
15 17 17 19 OL 11 12
OL
AM... D8S1179 D21S11 D18S51

100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
240
180
120
60
0
X Y 11 14 28
D19S433 TH01 FGA

100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
400
300
200
100
0
12 14 6 23
Figure 5.1 Low level partial DNA electropherogram showing degradation and/or inhibition of DNA and considerable drop-out.
All peaks except the X peak are below 200 RFU (inferring that they are below the stochastic threshold). OL = o
ff-ladder peak.
83
advanced and samples are collected (usually by swabbing) from areas that
exhibit no visible staining but might be expected to reveal biological mat-
ter deposited through handling, for example, knives and steering wheels.
Very small amounts of DNA (fewer than 100 pg or 0.0000000001 g) may
produce a DNA profile. But how far do we push these DNA testing tech-
niques so that we can be confident of reliable results?
The amplification kits commonly used in forensic laboratories usually
recommend a sensitivity threshold of at least 250 pg of template DNA. They
are not validated for quantities of DNA below that very small amount.
Issues surrounding the interpretation of DNA profiles using low level
analytical techniques such as low copy number (LCN) were brought to
the attention of the scientific community and the public domain after this
type of evidence was questioned by the presiding judge in the Omagh
bombing appeal in Belfast 2007 (Case 1). The accused was freed and
LCN DNA use was suspended in British courts for a period.
Case 1
The Omagh bombing occurred in 1998 and 29 people were
killed and 220 injured in a car bomb attack. Sean Hoey was
charged in 2005 after it was alleged that his DNA was found
on bomb timers collected through crime scene examination.
However, the LCN technique did not exist in 1998 and crime-
scene examiners did not necessarily follow the stringent
anti-contamination requirements for such examinations.
An independent report (Caddy et al., 2008) found that the
laboratory methods were robust and validated but confusion
remained in the interpretation of such profiles. The report
recommended that a DNA profile using low template DNA
techniques should be presented with clear caveats to juries in
criminal trials.
What is categorized as low level DNA may also be called low copy
number or low template DNA. What is important is that the laboratory
or the report defines the term used to describe low amounts of DNA and
explains the type of enhancement technique (such as increased amplifica-
tion) used, if any. The definitions should be stated in the body of the report
and/or any appendices. Gill et al. (2000) suggest insertion of a clause in
expert statements cautioning the court on the lack of interpretative infor-
mation such as transfer and persistence studies when determining the
value of low level DNA.
Chapter five: Partial profiles, low levels, and mixtures 85
While many laboratories perform testing with small amounts of DNA,

only a few formally conduct low template DNA case work with specific
enhanced detection protocols (Butler, 2012).
There has recently been a review commissioned by the Home Office
in the United Kingdom to address principles involved in the interpre-
tation of DNA profiles, especially those that are complex in some way
because the target material is at a low level, or degraded (Gill et al., 2012).
The review noted that the strength of evidence (to support a prosecution
or defense hypothesis) is likely to be maximized with the full conven-
tional DNA profile and minimized with the poorest interpretable low
level DNA profile.
5.2.2Stochastic effects
A forensic laboratory needs to determine at what point a detection
technique cannot deliver reliable results. Stochastic effects are random
sampling effects that may occur when a limited number of DNA target
molecules exist in a sample. Stochastic is derived from the Greek language
and describes systems whose behaviors are intrinsically nondeterministic
or random.
What happens with low amounts of DNA is that the PCR primers
used to amplify a specific region may not consistently find and hybridize
to the entire set of DNA molecules present in the amplification reaction.
With a heterozygous locus in which two alleles are present, unequal sam-
pling of the alleles can result in failure to detect one or both alleles. Loss
of a single allele is called drop-out while loss of both alleles is termed
locus drop-out. Other effects are unbalanced peak heights of paired
alleles and masking from a known or unknown contributor.
Drop-out arises when the allele carried by an individual contribut-
ing to a sample is not reported within the DNA profile obtained from the
sample. Drop-in occurs when trace amounts of DNA, for example, from
the crime scene environment or laboratory plasticware, generate one or
more spurious alleles in a profile. It is rare for drop-out and d rop-in to
occur with good quality samples not subjected to degradation or inhi-
bition, but d rop-in and d
rop-out become more likely as DNA amounts
decrease or environmental exposure increases.
Figure 5.1 shows both allele and whole locus drop-outs. Complete
locus d rop-out appears at D2S and at D18. This electropherogram may be
described as a partial DNA profile in a forensic report. All peak heights
appear to be below 200 RFU except for the X sex marker. In the authors
opinion this is a low level, suboptimal DNA profile that should not be
used for comparison purposes unless further analyses are performed to

obtain a better profile.
Stochastic variation is a fundamental physical law of the PCR ampli-
fication process when low amounts of DNA are examined. It manifests
as a fluctuation of results between replicate analyses. Thus it is possible
that amplifying the same extract twice can result in detection of different
alleles at a locus (Butler, 2012). Figure5.2 shows two electropherograms of
the same sample run twice. Note the different peaks at the D2 locus.
Since stochastic effects cannot be avoided when testing small amounts
of DNA, two approaches have been proposed: (1) stop testing or inter-
preting data before the stochastic realm is reached, or (2) try to limit the
impact by performing additional testing and following careful interpre-
tation guidelines based on validation studies. The second, or enhanced
interrogation, approach typically involves replicate testing and the devel-
opment of consensus profiles (Butler and Hill, 2010; Butler, 2012). Those
who advocate the second approach usually enhance method sensitivity in
order to get as much yield from a sample as possible. Careful validation
studies and appropriate interpretation guidelines are essential.
Recently, an approach has been advocated to completely eliminate
template thresholds from the definition because they represent an artifi-
cial cut-off for a continuous phenomenon. Instead, a risk assessment based
on peak heights of the DNA profile can be used to determine whether an
appropriate amount of DNA is present and stochastic factors are impact-
ing the typing (Gill and Buckleton, 2010). This paper describes a statistical
package that accounts for d rop-out and d
rop-in in probability terms.
This paper also critically examines the causes of the underlying con-
fusion about low template and low level DNA profile interpretations. The
biological model with replicate and consensus profiles was designed to
prevent misstatements about the strength of the evidence and present
suitable warnings about the limitations. The authors state that candor in
reporting should be applied. They believe that their statistical model is
the way forward, but in the absence of validated software implemented
in the specific laboratory issuing the report, the biological model can be
used with limitations expressed. This approach has also been recom-
mended by the Home Office (Gill et al., 2012). One of their principles is
that DNA interpretation methodology should incorporate a probabilistic
consideration of drop-out and additional alleles, such as drop-in, stutters,
gross contamination and additional contributors.
Fraser reviewIn 2010, in Victoria, Australia, a group of international
experts recommended changes in the state laboratorys DNA profile inter-
pretation methodologies. The chief commissioner of police suspended the
presentation of DNA evidence in court for six weeks; low level DNA pro-
files were of particular concern (Fraser et al., 2010).
Chapter five: Partial profiles, low levels, and mixtures
D3S1358 vWA D16S539 D2S1338
100 150 200 250 300 350
63
42
21
0
OL 17 16 10 24 26
OL 11
D3S1358 vWA D16S539 D2S1338

100 150 200 250 300 350
150
100
50
0
OL 17 16 10 23
11 24
Figure 5.2 Example of stochastic effects. R

e-analysis of the same sample revealed dfferent results at locus D2.
87
This action was taken due to the growing unease of the police and the
criminal justice community about the laboratorys methodologies. Three
experts from the United Kingdom and New Zealand were invited to
undertake a review of the laboratorys DNA interpretation practices. The
team was headed by Professor Jim Fraser and included John Buckleton
and Peter Gill, authors of numerous papers described in this book.
The review considered that the main events precipitating the con-
cern appeared to be (1) variations in statistics upon the implementation
of a new method designed to deal with peak drop-out due to low level or
degraded DNA, (2) dealing with artifacts in the profiles, and (3) inconsis-
tencies in interpretations of profiles by forensic scientists in the laboratory.
Very different statistical outcomes resulted from different methodologies
in different situations, and the differing outcomes raised concerns by the
police and prosecutors and generally resulted in a loss of confidence.
The Fraser team considered that removing professional judgment
from case managers was misguided because DNA profiling cannot be
deskilled to such an extent. They considered that the concern and loss of
confidence also arose from the organizational environment in the labora-
tory. The physical conditions were in need of significant improvement and
the cramped conditions and proximity of samples presented unnecessary
risks of contamination. The experts recommended the development of a
better appreciation of DNA interpretation practices throughout Australia
and internationally, broader engagement of staff, and the implementation
of a new DNA interpretation method. The findings of the Fraser review
can be applied to the consideration of DNA evidence by criminal justice
professionals in any jurisdiction.
Problems in interpretation of low level DNA profiles are drop-in and
drop-out, stutter, and unbalanced peak heights, in addition to masking from
a known contributor. There is a developed framework described above
for assessing such evidence based on likelihood ratios (LRs) that involve
drop-in and d rop-out probabilities (Gill and Buckleton, 2010). Ignoring a
discrepant locus or using the random man not excluded approach can
be systematically unfair to defendants. The LR depends strongly on the
assumed probability for drop-out, and ignoring the possibility of drop-in
is unfair to defendants.
5.2.3An interesting experiment

A criminology professor in the United States performed an experiment
involving inadvertent participants (DNA analysts) attending meetings
and conferences over a number of years (Thompson, 2009). They were
shown part of the electropherogram (Figure5.3) of an evidentiary sample
(presumed to be saliva) swabbed from the skin of a sexual assault victim.
D3S1358 vWA FGA
100
50
12 17 15 OL Allele ?
40 94 56 84 25
17 OL Allele ?
90 90 143
Defendant D3S1358 vWA FGA

Tom 17,17 15,17 25,25
Dick 12,17 15,17 20,25
Harry 14,17 15,17 20,25
Sally 12,17 15,15 20,22
Figure 5.3 Electropherogram of saliva sample and four suspect profiles. (Source:
Thompson, W.C. 2009. Law, Probability, and Risk, 8, 257276. With permission.)
The table at the bottom of the figure shows the alleles of four possible
defendants. Thompson stated that the choice of defendants who should
have been included or excluded as possible contributors to the evidentiary
sample was unclear.
At locus D3S1358 (D3) it must be determined whether the peak labeled
12 represents a true allele and, if so, whether it is associated with (from
the same contributor as) allele 17. At locus FGA, the determination must
be made whether the peak labeled OL allele? is a true allele or an arti-
fact. Another consideration is whether the electropherogram represents a
single source sample or a mixture.
At the first meeting, Thompson noted some uncertainty about the
inclusion of Defendant Tom who did not have the 12 peak at locus D3 and
asked how the participants could be sure that the true contributor did not
have genotype 12,17 at locus D3. Several analysts argued that the 12 peak
at D3 and the OL artifact should have been ignored because of the height
disparity between the 12 and 17 alleles and the poor morphology of the
12 allele. Of course, Defendant Tom was included.
At the next meeting, Thompson presented the evidentiary profile and
the defendant was presented as Dick rather than Tom. Thompson sug-
gested that the inclusion of Dick was problematic because of the uncer-
tainty whether the 12 peak at D3 was a true allele and because no 20 peak
had been detected at locus FGA. One analyst said the peak height dis-
parity was not an issue because these discrepancies are expected due to
stochastic effects. The OL allele was indeed an artifact that could have
masked an underlying 20 allele.
At a subsequent meeting, Thompson presented the defendant as
Harry. The analysts found no problem with this inclusion as the fail-
ure to detect the defendants allele 14 at locus D3 could easily be due to
allelic drop out and a 20 peak at locus FGA may have been masked by an
artifact. The peak labeled 12 at locus D3 was an obvious artifact. At that
point, Thompson wondered how much he needed to change the defen-
dant profile to get the analysts to agree that the defendant should have
been excluded.
A colleague of Thompsons, Dan Krane, also presented the case but
this time using a defendant profile labeled Sally. The analysts still insisted
that the defendant could not be excluded and invoked a mixture of two
contributors, one of whom had the 15 allele at locus vWA and the other
which had the 17 allele (Thompson, 2009).
The problems with this electropherogram are (1) it is a low level
DNA profile with all peaks below any laboratory stochastic threshold,
and (2) it presents the possibilities of d
rop-out, drop-in, and artifacts. In
the opinion of this author, this type of profile is suboptimal and not suit-
able for comparison with reference samples. It is far better to re-analyze
the DNA extract to obtain a better quality profile. If this is not possible, the
electropherogram should be marked as not interpretable due to quality
of profile.
5.2.4Enhancement techniques
Strategies exist for improving sensitivity in DNA analyses. An increased
number of PCR amplification cycles were first described in the late 1990s,
but they presented the possibilities of allele drop-out and drop-in and
increased risks of collection- and laboratory-based contamination. Later
methods increased the sensitivity of the injected product, improved
post-PCR purification, and reduced PCR volume (Butler, 2012).
5.2.5Improving reliability of results

There are two main areas where scientists aim to improve the reliability
of results obtained when potentially working with low amounts of DNA
(such as touch DNA from handled objects). First, they try to improve the
amount of DNA recovered at the collection and extraction stages. Second,
they limit the possibility of obtaining an incorrect answer by accounting
for stochastic effects in the analysis with a biological or statistical model.
5.2.5.1Biological (consensus) model

The moment of acceptance of stochastic effects will lead to difficulty in
inferring the contributing profiles of individuals from original or repli-
cate electropherograms. The biological model determines the profile via a
consensus strategy. The model was originally developed to facilitate the
reporting of DNA profiles that were subject to the phenomena of drop-out
and drop-in. It was later improved in order to facilitate reporting of low
template DNA profiles; the development of software solutions came later.
The statistical model discussed in the next section was concurrently
made available to check calculations provided by the biological model. The
biological model was devised to prevent misstatements of the strength of
the evidence and provide suitable caveats and warnings about the limita-
tions to be applied (Gill and Buckleton, 2010).
The replicate amplification strategy became the core feature of low
level DNA profiling (Butler, 2012). This concept was first published in 1996
(Taberlet et al., 1996). An allele was recorded only if it was observed at
least twice. The standard practice to date is to amplify two or three ali-
quots of a DNA extract (Gill et al., 2000; Caragine et al., 2009). Alleles that
occur more than once in the obtained profiles are designated. However,
amplification results from a single test may be unreliable due to the sto-
chastic effects (Butler, 2012), as noted above.
The Italian appellate court in the case of Amanda Knox and Raffaele
Sollecito (Case 7, Chapter2) held that the failure to perform two amplifi-
cations from the blade of the alleged weapon, despite the low quantity of
DNA, may be acceptable for initial investigative purposes but cannot
be accepted when the genetic tests form the basis for evidence of guilt
beyond any reasonable doubt (Hellmann, 2011).
The review commissioned by the Home Office (Gill et al., 2012) stated
that replication of a test for a compromised sample, although recom-
mended, was not compulsoryprovided that the interpretation can be
supported by a suitable statistical analysis. Splitting the sample into two
parts may be detrimental to interpretation and maximizing the sample
size for amplification will reduce ambiguity inherent in the DNA profile
of a compromised sample. A single analysis may provide the difference
between a test result that can be reported and one that cannot.
5.2.5.2Statistical (probabilistic) model

A strategy for interpreting low level DNA profiles and accounting for
stochastic variability was first introduced as a statistical model (Gill
et al., 2000). The statistical model can be used in two different ways: (1) to
develop a likelihood ratio per se or (2) to determine whether the consensus
approach is safe under the circumstances described. It attempts to assess

the probability of the replicates from all possible genotypes (Gill and
Buckleton, 2010).
The test of the strength of the evidence is assessed on a continuous
basis to formulate the LR. If alleles do not appear or are visualized just a
few times in multiple replicates, the LR is low.
The emerging literature describes the need for a move by the forensic
community toward more formal probabilistic models. The International
Society of Forensic Genetics (ISFG) commission supports the adoption of
these statistical type models (Gill et al., 2012). However, the slow adoption
of these models has been caused by complex concepts that are not rou-
tinely encountered in forensic science caseworkand are also difficult to
describe in a courtroom.
The essential feature of the classical or biological model is the pros-
ecution hypothesis of the binary determination of a match versus a
nonmatch that results in a probability of 1 or 0, respectively. However,
with the probabilistic model, the likelihood of the numerator can take any
value between 0 and 1 (Gill et al., 2012). Therefore the probability can be
described as a continuum, and this constitutes the fundamental differ-
ence in the two approaches.
If the DNA profiles only partially match between the crime and the
reference samples, uncertainty about the validity of the match is present
and the numerator cannot be described as 1.
Some s hort-cut calculations are used on occasion. The most common is
the 2p rule (Buckleton and Triggs, 2006). This rule is a shortcut method
to interpret partial profiles where drop-out has been invoked. For exam-
ple, only one allele peak may be seen at a locus in the crime profile but
the prosecution may invoke drop-out of a partner allele to accord with
their hypothesis. The weakness of this rule is that it does not take into
account the uncertainty of the match in the numerator of the prosecution
hypothesis: is there really a probability of 1 for a match with the suspect
profile? It has beenshown to be not conservativeespecially in the presence
of masking (Balding and Buckleton, 2009).
The statistical or probabilistic approach still requires a proper assess-
ment of the overall quality of the DNA profile in question and its suitabil-
ity for further analysis.
5.2.6Contamination
Contamination is a major issue when considering low level DNA. It is
possible to amplify the contaminant through enhancement techniques,
so it is imperative that a laboratory employs strict contamination mitiga-

tion measures. Crime scene collection procedures should also be subject
to scrutiny. Contamination was a major factor in the Omagh bombing
appeal and police collection procedures were criticized.
Similarly, the collection practices utilized to solve the murder of
Meredith Kercher were problematic (Case 7, Chapter2). The bra clasp from
the deceased was collected over a month and a half after the crime and
after the house had been subjected to several searches by nonscientific
examiners in the belief that, by then, all items of scientific relevance had
already been found. A photograph of the gloves of the scientific police
operatives taken when the clasp was collected showed traces of dirt
on the fingertips holding the clasp that could have been interpreted as
signs of previous soiling (Hellmann, 2011). The appeal court accepted the
theory of probable contamination of the bra clasp.
Every forensic laboratory today faces the question of low template or
low level DNA and must determine their validated guidelines. An aban-
donment of DNA testing at low levels is not generally considered practical
by the scientific community but a warning of the dangers of not under-
standing the potential for honest error and margins of error is vital. All
forensic scientists should communicate the limitations of their methods to
the criminal justice system.
The limitations are not confined to low template DNA profiles. A false
sense of security is a likely consequence when techniques are represented
by artificial divisions, that is, the results obtained from conventional pro-
files are interpreted using methods that do not follow the same cautions
applicable to low template DNA profiles (Gill and Buckleton, 2010). For
example, the Phantom of Heilbronn case in Germany (Himmelreich,
2009) involved widespread contamination that occurred in relation to con-
ventional DNA profiling (see Chapter8).
5.3DNA mixtures from two or more people

It is not unusual to observe DNA profiles containing contributions from
two or more people. These results may be expected, for example, from vag-
inal swabs from rape victims that carry mixtures composed of semen
from the rapist and vaginal cells from the victim. These types of samples
can be readily separated using different extraction techniques for sperm
and for skin cells. The problem in interpretation occurs most often with a
touch DNA mixture (because skin cells cannot be separated) or when the
body origins of the cells cannot be determined. Case 2 from the authors
files is an interesting example.
Case 2
A rape case was set for trial. The complainant in the matter
alleged she was asleep in the communal lounge room of a
boarding house when she awoke to find that the accused (who
also lived in the house) had raped her and she saw a used con-
dom on the floor. She said her tracksuit pants and underpants
had been removed while she slept.
The prosecution relied on the alleged finding of DNA from
the accused in a sample from the inner leg of the complainants
tracksuit pants and maintained that the sample was indica-
tive of semen. No semen was detected on any medical swabs
from the complainant. A review of the case notes showed that
samples from the left leg near the hem and from the inner knee
area of the tracksuit pants reacted positively to a screening
test for semen. However, semen could not be confirmed from
either sample and in fact the screening test showed slow reac-
tion times for both samples, indicating the possibility of false
positives.
A major DNA profile was obtained from a nonsperm frac-
tion from the hem area that matched the DNA of the accused;
no DNA was obtained from the sperm fraction. A mixture
DNA profile of at least three people was obtained from the
sperm fraction of the inner knee area including a partial DNA
profile that matched the corresponding components in the pro-
file of the accused with a statistic of 340,000 to 1.4 million.
A method of separating sperm cells from nonsperm cells
such as cellular material (for example the differential extrac-
tion method used in this case) does not confirm the presence
of sperm.
The hearing transcript also showed that the forensic scien-
tist in the case stated that a sperm fraction in a cellular separa-
tion did not confirm the presence of semen. No contribution of
the complainant was found on the clothing, and in fact a DNA
profile matched half of that of the complainant (possibly from
a parent). Further testing still could not confirm semen and the
prosecution decided not to lead any DNA evidence in the trial.
The differential extraction process for semen and sperm fractions is

explained in Chapter3. Case 2 emphasizes the importance of determining
whether DNA can be related to a specific body fluid, also described in
Chapter3.
The larger the number of contributors, the more complex the DNA
profile. If a sample contains material from four or more contributors, a
large portion of the population generally would be included in the pro-
file. Most laboratories do not attempt to perform interpretations of four or
more contributors to a mixed DNA profile unless a major contributor can
be determined.
As described previously, the peaks in DNA profiles represent alleles
that are genetically inherited from each parent. When the alleles from
each parent are of different sizes, the alleles will have different values.
Sometimes alleles of both parents are of the same size and the resulting
peak has the same values at each site (effectively doubling the height).
When mixtures contain DNA from two or more people, the assign-
ment of each allele to a particular individual becomes more difficult.
Computer programs are required to analyze a mixture of two or more
contributors without a clear major component.
When considering mixture DNA profiles, it is important to consider
the profile as a whole to determine the relative quality, quantity, and
number of contributors to the DNA. Knowledge of laboratory criteria is
required for understanding the assignment of the number of contributors,
the designation of a peak in a profile, and the conditions under which it is
appropriate to determine a statistic denoting evidential significance.
Case 3 illustrates what can happen if defense lawyers accept labora-
tory reports and expert testimony at face value without examining the
underlying scientific data (Thompson, 2003; Bromwich, 2007).
Case 3
In 2002, an inquiry by local television reporters into the opera-
tion of the Houston (Texas) police crime laboratory led to
reviews of several past cases by experts. One of these cases was
the conviction of Josiah Sutton for a 1998 abduction and rape
committed when he was just 16 years old.
A woman was abducted and raped at gunpoint in the back-
seat of her car and then dumped in a nearby field. She identified
Sutton and his friend as her two attackers. The DNA results
from the Houston laboratory noted a mixture of DNA from
two men, one of whom was Sutton. His friend was excluded.
DNA was the primary evidence against Sutton. However,
a defense expert found that although the laboratory deemed
Suttons DNA profile consistent with a mixture of alleles
found in some samples, the samples contained so many alleles
that thousands of people would also be consistent. The
defense expert excluded Sutton from contributing to the semen
stain on the backseat. If the unknown mans DNA on the back-

seat was in the DNA of the vaginal sample, Sutton could not
have been the other man. The jury never heard these possibili-
ties and took only two hours to convict Sutton.
What the jury heard was that every human being has a
unique DNA pattern and that Suttons pattern was found in the
sperm fraction of the vaginal swabs, on debris from combing of
the complainants pubic hair, and on her jeans.
The forensic report said that Suttons DNA profile would be
expected to occur in 1 in 694,000 members of the relevant popu-
lation. What the jury did not hear was the actual match of the
profile in the mixture (approximately 1 in 14). Apparently no
mixture statistic for Sutton was given. The inquiry led to DNA
retesting of the victims vaginal swab by another laboratory. The
laboratory found that the semen came from two men and elimi-
nated Sutton. Serious flaws, including incorrect and misleading
statistical calculations, were discovered in the original analysis.
The DNA section of the Houston crime laboratory was shut
down in 2002, reopened in 2006, and again shut down in 2008
and also subjected to audit and review. Sutton was exonerated
in 2004 after serving more than 4 years of a 25-year sentence.
The real perpetrator was found in 2006 through a DNA CODIS
database match. He pleaded guilty and was sentenced to
10 years. The other perpetrator apparently died during impris-
onment for other charges.
The interpretation of a mixed DNA profile is relatively simple when

the following criteria apply (Word, 2011):
The DNA comes from only two sources.

The two sources are unrelated and have no or few shared alleles.
The ratio of the amount of DNA contributed by each of the two
sources is adequate for interpretation of both sources.
The appropriate amount of DNA was amplified and the alleles for
both sources exceed the analytical threshold of the laboratory.
No degradation, inhibition, or primer variants are present to affect
peak heights and the apparent DNA ratio.
All stutter peaks and other artifacts are below the analytical thresh-
old or clearly distinguishable as artifacts.
Significant alteration to any of the above parameters will likely make mix-
ture interpretation more complex; a combination of several alterations
generally confounds an interpretation significantly (Word, 2011).
According to Guideline 3.6.1 of the Scientific Working Group on DNA

Analysis Methods (SWGDAM, 2010; Mixture Interpretation Workshop,
2011), an analyzing laboratory must establish guidelines to ensure that,
to the extent possible, DNA typing results from evidentiary samples are
interpreted before comparison with any known samples other than those
of assumed contributors.
Masking of peaks or sharing of alleles is a common result from mix-
tures because a contributor reference DNA profile from a mixture may
have a particular allele or value at a particular locus that is the same as
that from another contributor profile. The contributions may be additive if
they are similar in amount or they may not be observed if one contributor
has donated less DNA than the other(s). It is necessary to make an assess-
ment in relation to the heterozygote balance and mixture proportion (Gill
et al., 2006).
In some cases DNA from one contributor to a mixed DNA profile may
be present in a larger amount than DNA from another contributor. This
component is sometimes referred to as originating from the major con-
tributor and may be interpreted as a single source profile. In other cases,
none of the contributors to a mixed DNA profile can be inferred (Kelly
et. al, 2012). Laboratories have protocols for determining whether major
and minor contributions are present in a mixtureoften when the minor
contribution is less than 30% of the total. The calculations are then rela-
tively simple. However, computer programs are needed for unresolvable
mixtures, for example, when two contributors cannot be distinguished as
major and minor.
In mixture calculations, the concepts of restricted and unrestricted come
into play. In a restricted calculation, the relative peak heights at each locus
are taken into account when pairing the alleles for the calculation. In an
unrestricted calculation, all of the possible combinations of the alleles are
deemed possible and are thus used in the calculation.
The likelihood ratio calculation method in mixture DNA profiles
where the peak heights and areas are not taken into account is called the
unrestricted combinatorial method (Evett et al., 1991; Weir et al., 1997).
This method examines all possible sets of genotypes consistent with
the alternative hypotheses (prosecution and defense) and utilizes uni-
form assumptions such as the number of contributors across the loci.
This method does not account for the possibility of drop-out. There is a
significant risk that the LR will be significantly nonconservative (Kelly
et al., 2012).
Better statistical models are now becoming available. These include
semi-continuous models like LoComatioN (Gill et al., 2007) or fully con-
tinuous models like TrueAllele (Perlin et al., 2011).
5.4Mixture interpretation steps

These procedures are modified from Clayton et al.,1998 and the SWGDAM
guidelines (2010).
Step 1: Identify the presence of a mixtureIf more than two alleles
appear at a locus, the presence of a mixture may be inferred. However,
extra peaks may be present because of stutters and other artifacts. Allele
symmetry may arise because of shared alleles and lead to masking; as a
result the profile will appear unbalanced.
Step 2: Designate allelic peaksThe peaks are designated as true
alleles or as artifacts such as stutter, dye blobs, pull-up, and so on.
Step 3: Identify the number of contributors in the mixture
The number of alleles per locus, circumstances of the case, and pos-
sibility of related contributors factor into the determination of number
of contributors.
Step 4: Estimate the mixture proportion , i.e., estimate the ratios
of the individuals contributing to the mixtureIt may be possible to
separate the mixture into major and minor components. In analyzing a
mixture, the ratio or proportion of each contributor is approximately pre-
served throughout the mixture at each locus.
Step 5: Consider all possible genotype combinationsAll combina-
tions of the unrestricted combinatorial list of genotypes are considered
in relation to the mixture proportion and the heterozygote balance across
all loci.
Step 6: Compare reference samplesIt is important that the previ-
ous steps take place without considering the reference samples to demon-
strably avoid the possibility of bias.
It may be necessary to consider different propositions at various stages
of the analysis. Ultimately the court will decide those that are relevant for
consideration. The prosecution and the defense both seek to maximize
their respective probabilities of the evidence profile. There is no reason
not to evaluate multiple pairs of propositions (Buckleton, 2005). The DNA
result itself may indicate that different explanations are possible. One
common misconception is that the numbers of contributors under the
prosecution and defense hypotheses should be the same but there is no
reason for this to be so.
Both parties should confer with their respective forensic scientists to
establish their hypotheses. The smallest numbers of unknown contribu-
tors are usually proposed in order to explain the evidence and maximize
the desired likelihood. However, it is sometimes wise to denote different
options for different numbers of contributors so that a court delay for cal-
culations is not required.
5.5Low template mixtures

Stringent criteria should apply to the interpretation of mixtures and defi-
nition of peaks as true peaks. These criteria are difficult to meet in tests of
touch DNA, as very small amounts of DNA may be recovered. In cases
of mixtures, one of the contributors may be considered as low template
DNA even though the total amount is 1 ng (say a 1:9 mixture). Thus issues
of low template DNA must be considered.
Incorporation of an assessment of the probability of allele drop-in and
drop-out for low level DNA mixtures has been recommended. Another
recommendation is that more empirical, quantitative data should be gen-
erated on the effects of low levels of DNA within mixtures. It is known
that the stochastic nature of low template amplification renders any peak
height threshold inaccurate below about 267 relative fluorescent units
(RFUs). Therefore, caution has been urged for mixture interpretation
when only trace amounts of DNA from one or more of the contributors
are present (van Oorschot et al., 2010).
Whenever d rop-out is a possibility, no meaningful exclusion prob-
ability can be calculated for the full profile. A random match probabil-
ity (RMP) and a likelihood ratio (LR) approach may be extended to deal
with situations where drop-out is possible and no nonconcordant alleles
are present. A n onconcordant allele is present in the person of interest
but not visualized in the electropherogram. The LR approach, but not the
RMP, may be further extended to handle situations where nonconcordant
alleles exist.
In cases where drop-out is invoked to sustain the prosecution case
(not all the accuseds alleles are present in the crime DNA so there is
nonconcordance of alleles), estimation of d rop-out probabilities cannot be
avoided (Balding and Buckleton, 2009).
For mixed DNA profiles of low template DNA exhibiting stochas-
tic effects, the calculation of LR may proceed via either a binary, semi-
continuous, or full continuous method (Kelly et al., 2012). The binary
method treats alleles as present or absent. The semi-continuous method
assigns a probability to the events of d rop-out or no d rop-out but still
treats alleles as present or absent. Fully continuous methods deal with the
probabilities of stochastic events (like d rop-out) based on the heights of
the peaks visualized at a locus. No modification of the binary method can
deal with a nonconcordant allele in a comprehensive manner (Buckleton
and Triggs, 2006).
Drop-out must not be a possibility for the unconstrained combina-
torial method. Methods have been described for complex mixtures in
which d rop-out is possible but there are no n onconcordant alleles (Kelly
et al., 2012).
5.6Complex mixtures
If a complex or indistinguishable mixture involves at least three individu-
als and no clear major contributor appears, L R-calculated yields limited
evidential value. Sometimes a sample includes so many common alleles
that few people are excluded, as in Case 4 from the authors files.
Case 4
An armed robbery by two masked men was committed at the
home of a woman and her two young children. Accused B
pleaded guilty and said Accused A was the main perpetrator.
Touch and wearer DNA samples from the handles of a sports
bag and the inner armpits and collar of a black suit jacket left
in the backyard of the home were obtained by tape lifts. A sta-
tistical analysis found that Accused A was not excluded as a
contributor to the DNA detected from the two items.
LR statistics were performed considering two propositions
for the DNA from the handles of the sports bag: (1) the DNA
originated from Accused A, Accused B, and one other person
chosen at random from the population or (2) the DNA origi-
nated from three other people chosen at random. The LR was
determined to be 0.93. In other words, it was estimated to be
1.1 times less likely that the first proposition was true than if
the second proposition was true.
Statistics were also performed considering two proposi-
tions for the DNA from the suit jacket: (1) it originated from
Accused A and two other people chosen at random or (2) it
originated from three other people chosen at random. The LR
was 0.4. In other words, it was estimated to be 2.5 times less
likely that the first proposition was true than if the second
proposition was true.
There was more support for the proposition that the DNA
from both the handles of the sports bag and the suit jacket
came from three other people chosen at random from the popula-
tion rather than for the proposition that the DNA came from
Accused A and two other people (B and one unknown for the
sports bag, or two unknowns for the jacket). Accused A was
found not guilty after the jury deliberated less than 30 minutes.
If a person of interest is included in a mixture and exhibits some

common alleles, it may be harder to find two other individuals with the
remaining (possibly rarer) alleles than it is to find three random people
who fit all the alleles in the profile. The inclusion of the person of inter-
est is not surprising as almost everyone is included but less included
than average.
When a complex profile shows obvious stochastic effects, it is consid-
ered uninterpretable. The mixture classification scheme of the German
Stain Commission (Schneider et al., 2009) denotes a mixture profile
without major contributor(s) and with evidence for stochastic effects as
uninterpretable. No method for interpretation currently exists for com-
plex mixtures involving four or more persons, and thus such mixtures are
not suitable for comparison to a person of interest.
Results from simulation studies of a four-person mixture (Paoletti
et al., 2005) showed that 0.02% would show four or fewer alleles and that
76.35% would show six or fewer alleles using the U.S. 13-locus CODIS
system. Consequently, more than 70% of four-person mixtures would
not be recognized as involving four persons based on allele counts. The
studies showed a t hree-person mixture would be incorrectly designated
a t wo-person mixture in a small percentage of cases. These studies also
show the inherent problems of interpreting complex mixtures from three
or more contributors.
References
Balding, D. and Buckleton, J. 2009. Interpreting low template DNA profiles.
Forensic Science International: Genetics, 4, 110.
Bromwich, M. 2007. Final report of the independent investigation for the Houston
Police Department Crime Laboratory and Property Room. http://www.
hpdlabinvestigation.org/reports
Buckleton, J. 2005. A framework for interpreting evidence. In Buckleton, J. et al., Eds.,
Forensic DNA Evidence Interpretation. Boca Raton, FL: CRC Press, pp.2763.
Buckleton, J. and Triggs, C.M. 2006. Is the 2p rule always conservative? Forensic
Butler, J. and Hill, C.R. 2010. Scientific issues with analysis of low amounts of DNA.
Profiles in DNA, 13. http://www.promega.com/profiles/1301/1301_02.html
Caddy, B., Taylor, G.R., and Linacre, A.M.T. 2008. A review of the science of low
template DNA analysis. Home Office Forensic Regulation Unit. http://
police.homeoffice.gov.uk/publications/operational-policing/
Caragine, T., Mikulasovich, R., Tamariz, J. et al. 2009. Validation of testing and
interpretation protocols for low template DNA samples using AmpFLSTR
Identifiler. Croatian Medical Journal, 50, 250267.
Clayton, T., Whitaker, J.P., Sparkes, R.L. et al. 1998. Analysis and interpretation of
mixed forensic stains using DNA STR profiling. Forensic Science International,
91, 5570.
Crown Prosecution Service. 2010. Summary of national DNA database from the
prosecution perspective. http://www.cps.gov.uk/legal
Evett, I.W., Buffery, C., Willot, G. et al. 1991. A guide to interpreting single locus
profiles of DNA mixtures in forensic cases. Journal of Forensic Science Society,
31, 4147.
Fraser, J., Buckleton, J., and Gill, P. 2010. Review of DNA reporting practices by
Victoria Police Forensic Services Division. http://www.vicpolicenews.com.
au/images/stories/news/feature_story/victoria%20police%20forensic%20
services%20review%20%20report%20%20april%202010.pdf
Gill, P., Guiness, J., and Iveson, S. 2012. The interpretation of DNA evidence
(including low template DNA). Forensic Science Regulator, Home Office
Forensic Regulation Unit, Crown Copyright, July. http//homeoffice.gov.uk.
Gill, P., Gusmao, L., Haned, H. et al. 2012. DNA Commission of the International
Society of Forensic Genetics: Recommendations on the evaluation of STR
typing results that may include drop-out and/or drop-in using probabilistic
methods. Forensic Science International: Genetics, 6, 679688.
Gill, P., Brenner, C.H., Buckleton, J. et al. 2006. DNA Commission of the International
Society of Forensic Genetics: Recommendations on the interpretation of mix-
tures. Forensic Science International, 160, 90101.
Gill, P., Kirkham, A., Curran, J. 2007. LoComatioN: A software tool for the analysis
of low copy number DNA profiles. Forensic Science International, 166, 128138.
Gill, P., Whitaker, J., Flaxman, C. et al. 2000. An investigation of the rigor of inter-
pretation rules for STRs derived from less than 100 pg of DNA. Forensic
Hellmann, P. 2011. The H ellmann- Zanetti report on the acquittal of Amanda
Knox and Raffaele Sollecito. December 16. http://www.hellmannreport.
wordpress.com
Himmelreich, C. 2009. Germanys phantom serial killer: A DNA blunder. Time
(Berlin).
Kelly, H., Bright, J., Curran, J. et al. 2012. The interpretation of low level DNA mix-
tures. Forensic Science International: Genetics, 6, 191197.
Mixture Interpretation Workshop. 2011. 22nd International Symposium on Human
Identification, October 3. http://www.cstl.nist.gov/biotech/strbase/training.
htm
Paoletti, D.R., Doom, T.E., Krane, C.M. et al. 2005. Empirical analysis of the STR
profiles resulting from conceptual mixtures. Journal of Forensic Sciences, 50,
13611366.
Perlin, M.W., Leger, M.M., Spencer, C.E. et al. 2011. Validating TrueAllele DNA
mixture interpretation. http://www.cybgen.com/documents/publications/
PLSSABD2010.pdf
Schneider, P.M., Fimmers, R., Keil, W. et al. 2009. German Stain Commission:
Recommendations for the interpretation of mixed stains. International Journal
of Legal Medicine, 123, 15.
SWGDAM (Scientific Working Group on DNA Analysis Methods). 2010.
Interpretation guidelines for autosomal STR typing by forensic DNA test-
ing laboratories. FBI website: http://www.fbi.gov/about-us/lab/codis/
swgdam.pdf
Taberlet, P., Griffin, S., Goossens, B. et al. 1996. Reliable genotyping of samples with
very low DNA quantities using PCR. Nucleic Acids Research, 24, 31893194.
sharpshooter fallacy in forensic DNA interpretation. Law, Probability, and
Risk, 8, 257276.
Thompson, W.C. 2003. Review of DNA evidence in State of Texas v. Josiah Sutton.
February 6. http://www.scientific.org/archive/Thompson%20Report.pdf
van Oorshot, R., Ballantyne, K., and Mitchell, R. 2010. Forensic trace DNA: A
review. Investigative Genetics, 1, 14.
Weir, B.S., Triggs, C.M., Starling, L. et al. 1997. Interpreting DNA mixtures. Journal
of Forensic Sciences, 42, 213222.
Word, C.J. 2011. Mixture interpretation: Why is it sometimes so hard?
http://www.promega.com/resources/articles/profiles-in-dna/2011
chapter six
Y-STR profiling
A number of DNA techniques use different sequences from the autosomal
STR DNA profiling methods described in the previous four chapters. This
chapter will discuss Y chromosome short tandem repeat (Y-STR) profiling
in humans. Chapter7 covers mitochondrial DNA in humans and animals
and more recent techniques such as familial DNA typing.
6.1Introduction
Continuing scientific discoveries leading to improvements and expanded
applications of DNA are regularly reported in the (vast) literature and it
may be a challenge for a legal practitioner to keep abreast of advances in
the field. Details of the analysis of a forensic sample and consideration of
the various DNA techniques employed should be described in a foren-
sic science report as a guide to legal practitioners and other nonscien-
tific reviewers.
Legal practitioners should be aware of other discriminating tech-
niques beside nuclear autosomal DNA profiling (discussed in the previ-
ous four chapters) that analyze human DNA for purposes of identification
in criminal cases. These additional techniques utilize different prim-
ers and/or sequences. Examples are the Minifiler STR that uses shorter
tandem repeats, mitochondrial DNA profiling, and Y -
STR profiling.
Mitochondrial DNA and Y-STR testing are genealogical techniques. The
results from living individuals are often compared to historic populations
or individuals. These tests are based on haplotypes (or complete sequence
types) and are less discriminatory than autosomal STR profiling.
The statistics for a Y chromosomal or mitochondrial DNA haplo-
type are different from those for STR autosomal DNA. The haplotypes
must be treated mathematically as a single indivisible (atomic) trait. Thus,
unlike traditional DNA methods that examine several traits that are
approximately independent of each other, no multiplication of probabili-
ties is possible with haplotypes. Therefore it is vital to have a sound fun-
damental understanding of atomic trait matching probabilities to make a
reasonable assessment of the strength of identification evidence if these
methods are used.
Y-STR profiling analyzes variations on the male (Y) chromosome in
nuclear DNA. This technique can be used when autosomal (nuclear) DNA
105
typing is unsuccessful on a crime sample that contains male material and

a possible donor needs to be determined. It is also particularly useful in
cases where the DNA recovered from an item contains a mixture of male
and female DNA. Since females possess only X chromosomes and no
Y chromosome, this difference is exploited in order to target only the male
DNA in a mixture containing male and female DNA. By contrast, males
and females both inherit mitochondria from their mothers (see Chapter7).
6.2Benefits
-STR typing may achieve profiles for male DNA (1) in samples contain-
Y
ing low levels of male DNA and high background levels of female DNA,
(2) in mixtures in which the female portion is present in overwhelming
quantities compared to the male portion, (3) where there are multiple
male contributors, and (4) in extended interval postcoital cervicovaginal
samples. The following is a brief summary of situations in which Y -STR
profiling may be worth considering (Roewer, 2009; Jobling and Gill, 2004):
Mixed stains in which the proportion of female DNA is higher than

the male DNA present (frequently observed in vaginal swabs col-
lected after sexual intercourse)
Cases of alleged sexual assault in which tests for seminal fluid or
sperm are negative
Sexual assault cases in which the evidence is positive for semen, but
no DNA foreign to the victim can be detected or potential male allele
levels are below the threshold for autosomal STR detection
Sexual assault cases in which the evidence is amylase-positive
(saliva-presumptive) and a male-and-female mixture is expected
(e.g., traces of kisses or bites)
Cases with very old semen stains in which most sperm cells are sus-
pected to be degraded and differential lysis is unsuccessful or risky
Sexual assault cases requiring screening of a large number of semen
or other stains
Cases requiring determination of the number of male donors in a stain
Cases in which the evidence is expected to include cells of a male
perpetrator (for example, underneath a female fingernail where
male biological material may accumulate after a violent attack)
Cases requiring determination of the patrilinear relationship of a
stain donor
Cases in which the stain donors population of origin must be inferred
Another use is profiling in a case involving a very small number of male

cells present at a rape scene because the rapist is azoospermic (lacks sperm)
Chapter six: Y-STR profiling 107
or oligospermic (has a low sperm count). Y-specific typing may be effec-

tive even if a vasectomized or azoospermic male leaves no sperm after
coitus and the sample presents a 4000-fold excess of female DNA (Jobling
and Gill, 2004).
Autosomal STR analysis may not be useful if a sample contains an
admixture of body fluids other than semen (saliva and saliva, saliva and
vaginal secretion, fingernail scrapings revealing cells from a female vic-
tim and male perpetrator). It is not possible to use differential extraction
to separate the male and female cells in such samples with current tech-
nology. The male component is often not detectable with the autosomal
STR multiplex systems routinely used.
Autosomal STR analysis may also fail with s emen-containing samples
in which very low copy numbers of sperm are present, or extremely frag-
ile postcoital examples are taken after an extended interval (e.g., 48 hours
after crime). Differential extraction of these samples may yield no pro-
file from a male donor due to a combination of premature lysis of the
cellular constituents of the sperm into the nonsperm fraction and sperm
loss during the physical manipulations required for the isolation process.
Therefore, Y -STR profiles that target only the male fraction eliminate the
need for a differential extraction process and lessen the potential to lose
the very small amounts of male DNA that may be present (Mayntz-Press
et al., 2008).
Y-STR analysis presents several additional benefits in forensic case-
work. It allows the easy determination of the number of male contributors
in a mixture. The profiles are hemizygous (one allele found at most loci)
with the exception of a small number of multicopy loci). Multiple alleles at
single copy loci clearly indicate the number of male contributors.
Shortly after the characterization and evaluation of the first
Y-chromosomal STR polymorphism, its usefulness in crime casework
was demonstrated when a mixed stain from a vaginal swab of a raped
and murdered female victim was resolved by Y-STR analysis and a falsely
convicted male was excluded in 1992 (Roewer et al., 1992). Case 1 involved
a request for retrial in Japan in 1998. The request was denied after 25-year-
old vaginal swabs from two murder and rape cases were subjected to
Y-STR testing and matched the stains to each other and the defendant
(Honda et al., 1999).
Case 1
A retrial of a condemned criminal whose capital punishment
had been suspended was requested by the Sapporo High Court
in Japan. From 1972 to 1973, two successive rape and murder
crimes and another rape occurred in Hokkaido in northern
Japan. Two years later, a suspect was arrested. He confessed to

the crimes during the trial and was sentenced to life imprison-
ment. However, he later insisted on his innocence. A second
trial found that he was guilty and he was sentenced to death.
The defense then appealed to the Supreme Court but
the capital punishment was not overturned. After the judg-
ment of the Supreme Court, the defense demanded a retrial
on the ground that the judgment based on a confession was
unsatisfactorythe original ABO blood grouping evidence
was inconclusive. DNA was extracted from mixed seminal
and vaginal secretions collected 25 years earlier from the two
raped and murdered victims. Y -STR profiling was performed
on the samples and four Y-STR types were found identical to
those of the accused. The High Court accepted the results and
refused the retrial request in 1998.
Y-STR profiling may be useful in paternity testing, especially if the

father is no longer alive. An example is the historically controversial
case of Thomas Jefferson, the third president of the United States, who
was thought to be the father of a child of Sally Hemings, one of his slaves.
Jefferson had no sons so the descendants of his paternal uncle were sub-
jected to Y -STR profiling. The Y
-STR haplotypes were compared with
male line descendants of the last son of Hemings. The results showed the
same haplotype and supported the proposition that Jefferson could have
been the father of Hemings child (Foster et al., 1998).
In Case 2, the Y-STR typing of a large population was used to elimi-
nate suspects and then autosomal STR profiling was used to identify a
serial rapist and murderer in Poland in the early 2000s (Detlaff-Kakol and
Pawlowski, 2002).
Case 2
A man committed at least 14 rapes in Poland since 1996 and
murdered a 22-year-old woman in 2000. DNA profiles obtained
from semen stains left at the crime scenes indicated that one
male committed all the rapes. The Y chromosome haplotype
obtained from the DNA in the semen stains was used to elimi-
nate 421 suspects.
One man exhibited a DNA profile identical at all Y chro-
mosome STR loci analyzed and possessed common alleles
in 9 of 10 autosomal loci. These findings strongly suggested
that the rapist and the man who exhibited the identical Y-STR
profile were closely related males. Analysis of reference DNA
obtained from the mans brother revealed an identical autoso-
mal STR profile to those identified at the crime scenes.
The U.S. case of A.B. Butler demonstrates the benefit of this method in
excluding an individual convicted years earlier (Innocence Project). The
accused was sentenced to 99 years and imprisoned in Texas in 1983 for kid-
napping a woman from a parking lot and then raping her. The biological
evidence was not tested until 1999 and autosomal DNA analysis did not
yield conclusive results. Y
-STR profiling had just been implemented in the
New York Medical Examiners Office and the Texas evidence was then
sent there. The results excluded Butler as the source of the semen from the
rape kit. Butler was released in 2000 after serving 16 years in prison and
was pardoned and compensated.
6.3Theory
The Y chromosome is paternally inherited and the profile is called a hap-
lotype. These haplotypes are less diverse than the genotypes utilized
in autosomal STR profiling containing equivalent numbers of markers.
Patrilineal relatives (brothers, father, sons, paternal uncles) of a particular
male will share a haplotype and this factor must be considered in any
evidential analysis.
The principal weakness of Y chromosome STR analysis is that even
when a crime sample matches the profile of a suspect, patrilineal rela-
tives of the suspect cannot be excluded as donors of the stain. Hence, in
contrast to autosomal STRs, access to reference databases representing the
variance and relatedness of haplotypes within local populations is crucial
for interpreting Y -STR matches.
The Y chromosome in humans is approximately 40 million base pairs
long and contains just 78 genes. The sex-determining region on the Y chro-
mosome encodes a protein that triggers the development of the testes and,
through an extended hormonal pathway, causes a developing fetus to
become male. Most of the Y chromosome is n on-recombining and passes
unchanged from father to son except when mutations occur. This lack of
recombination may be the reason why the Y chromosome reveals rela-
tively few genes.
Many Y -STR loci have been described in the literature (Butler, 2006).
The basic repeats for most Y-STR loci (as in autosomal STR profiling) are
tetranucleotides (four-base pairs).
6.4Statistics
6.4.1Frequency estimates of Y-STR haplotypes
-STR profiling is very useful for exclusionary purposes because the
Y
result is unequivocal without the need to provide a statistical weight.
If the Y
-STR evidence is inclusionary, a statistical weight must be
applied. The Y -STR loci are inherited from father to son as a single unit,
virtually unchanged in each generation except for occasional mutations.
Therefore, the haplotype of a man should be the same as his biological
brothers, sons, and all other males along the paternal lineage. This hap-
loidy and patrilineal inheritance complicate the interpretation of Y -STR
haplotype matches because male relatives share identical Y-STR profiles
for several generations.
Calculating statistics for Y -STR profiles is considerably different from
developing statistics for autosomal DNA profiles. All Y-STRs show link-
age to the Y chromosome so that multiplying frequencies cannot be used
to determine the frequency of a haplotype.
Linkage and smaller effective population size contribute to
population-specific distributions that are affected by genetic drift and
geographic differentiation. Population substructure effects have been
shown to be more substantial for Y loci compared with observations for
autosomal STR loci. Large databases of haplotypes must be maintained
(typically by race or ethnic group) and the databases are then searched for
haplotypes that match the haplotype of interest.
Haplotype frequencies observed in or extrapolated from these data-
bases often range between 1 in 1000 and 1 in 100,000, much lower than
the 1 in 1 billion typically cited in forensic reports for autosomal DNA
profiling. The Y Chromosome Haplotype Reference Database (YHRD;
www.yrhd.org) is an online facility designed to store Y chromosome
haplotypes from global populations and replaces three separate data-
base collections of European, Asian, and United States chromosomes
(Willuweit and Roewer, 2007). As of February 2012, it contained over
100,000 haplotypes from more than 750 populations in 109 countries.
The Y chromosome has a n onrandom distribution among global
populations due to a practice known as patrilocality (the female moves
to the males birthplace after marriage). Therefore, the priority for popu-
lation sampling should not be sample size alone but should also include
a good representation of the spectrum of p opulation-specific haplotypes.
When sampled properly, even populations such as the Europeans, formerly
regarded as sufficiently homogeneous for purposes of forensic genet-
ics, appear genetically subdivided into distinct Y chromosomal clusters
formed and maintained by recent demographic events (Roewer et al., 2005).
It is incumbent upon forensic scientists to assess the effects of popu-

lation substructure and employ statistical approaches that address those
effects for the relevant populations. The limitations of the specific data-
base used and the meaning of any match found should be enunciated
clearly in forensic reports.
6.4.2Meaning of Y-STR match

A conservative statement for a Y
-STR match report may be:
The Y-STR profile of the crime sample matches the Y -STR pro-
file of the suspect (at xxx number of loci examined). Therefore,
we cannot exclude the suspect as being the donor of the crime
sample. In addition, we cannot exclude all patrilineal related male
relatives and an unknown number of unrelated males as donors
of the crime sample.
Case 3 from the authors files illustrates that DNA evidence can still be
obtained even though initial attempts with autosomal DNA profiling
were unfruitful due to a lack of spermatozoa.
Case 3
The estranged partner of a woman was alleged to have burst
into her home and raped her vaginally. He was also alleged
to have put an axe handle in her vagina although it was never
located. A single spermatozoon was determined from a high
vaginal swab but no spermatozoa were found on any of the
other swabs. No autosomal STR profiles were obtained from
the medical swabs.
A Y-STR analysis was performed and a result obtained
from the high vaginal swab (cellular fraction) and a labial swab,
both with expected frequencies of 1 in 163 in the database (the
haplotype was observed once in the database) that contained
profiles of 1,079 individuals. The accused pleaded guilty at the
beginning of the trial due to other issues so that the probity of
the DNA evidence was not tested.
Two current approaches are utilized to evaluate the probability of a

coincidental match between two Y-STR haplotypes if a frequency estimate
is required. The most common method that has been used for many years
with mitochondrial DNA profiling is the counting method (Gill et al.,
2001). The other is the haplotype surveying method (Roewer et al., 2005)
which is a Bayesian approach that attempts to extract more information
from the structure of Y -STR haplotype databases than does the count-
ing method.
An estimate of the frequency of the haplotype in a population is not
possible by just calculating the number of matching profiles divided by
the total number of profiles in a database. On many occasions, the specific
haplotype has not been observed in the database (a null frequency), often
due to the limited database size (see Case 3).
A conservative bound is thus placed on the estimate to correct for
possible sampling error. The confidence interval allows for a measure of
the amount of confidence that may be placed on a value lying between
two specified limits (the interval). One can calculate the upper bound of
the confidence interval and this value can be used to convey, with a high
degree of confidence, that the rarity of the evidence Y-haplotype among
unrelated individuals in a given population is less than the upper bound
of the estimate. The assumption of a normal distribution may not apply
for Y-STR haplotype frequency estimates, but assuming normality will
provide a conservative upper bound estimate.
The Scientific Working Group on DNA Analysis Methods (SWGDAM)
states that the use of the counting method that incorporates the
upper-bound estimate of the count proportion offers an appropriate and con-
servative statistical approach to evaluating the probative value of a match
(2009). The following calculations are based on SWGDAM guidelines.
EXAMPLE 1
For a haplotype that is not observed in a database, the following
formula is used to calculate the upper 95% confidence interval and
serves as a correction for sampling uncertainty:
1 (0.0.5) to the power of 1/n
where n = size of the database. Assume that n = 2000, and there have
been 0 observations previously in the database. Then 1 (0.05) to the
power of 1/2000 = 1 0.9985032 = 0.0014967, or approximately 1 in 668.
EXAMPLE 2
For a haplotype observed previously within a database, the calcula-
tion is:
Upper bound = p + 1.96 ((p)(1p))/n
Lower bound = p 1.96(p)(1p)/n

where p = x/n, n = database size and x = number of observations of the

haplotype in the database. Assume Mr.A has a haplotype that was
observed five times previously (x = 5) in a database of n = 2000:
Upper bound = 5/2000 + 1.96((5/2000)(15/2000))/2000

= 0.0025 + 1.96(.00111663)
= 0.0046886 (approximately 1 in 213)
A report should indicate that, Mr.A could have contributed to the male
source of the DNA detected. In addition, all male relatives on the pater-
nal line and approximately 1 in 210 unrelated males cannot be excluded.
Appropriate caveats on the limitations of the database used should also
be explained.
6.5Number of male contributors to Y-STR profile

When more than one male contributes to a mixture, the individual male
haplotypes may be difficult to distinguish unless they are present in dif-
ferent quantities. Y-STR testing may still be useful for determining the
number of male contributors in a mixture. Single copy Y-STR loci usu-
ally produce a single amplicon in single source samples. Multiple peaks
observed at such a locus may suggest two or more male contributors.
While this is true in most cases, many regions of the Y chromosome are
duplicated or even triplicated in some individuals and this can complicate
mixture interpretation.
6.6Determining mixture ratios

It is possible to determine a major or minor contributor to a mixed Y-STR
profile. Unlike the autosomal STR assays, the peak heights across the range
of molecular sizes do not remain relatively constant (maintain roughly
the same peak height at different loci in the same dye) in Y -STR analysis.
Thus, some of the smaller loci may disappear before the larger loci.
For cases in which more than one haploid profile is detectable in a
stain (for example, in a gang rape case), a likelihood calculation for vary-
ing numbers of known and unknown male contributors has been devised
(Wolf et al., 2005). A prerequisite for such calculations is again the use of
large Y-STR haplotype population databases to retrieve frequencies of the
haplotype profiles detected in the trace.
However, two limitations of Y-STR mixture analyses must be
addressed: (1) the need to determine frequencies that apply for all possible
haplotypes that contribute to a mixture; and (2) u

ser-friendly interpreta-
tion algorithms for Y
-STR evidence profiles.
6.7Combining statistics from autosomal

and Y-STR profiling
No currently accepted method exists for combining autosomal STR and
Y-STR profiling. The different natures of underlying population structures
infer that a combination of the information obtained from lineage genetic
markers such as the Y chromosome (or mitochondrial DNA) with data
resulting from meiotically recombining loci (autosomes) into a single like-
lihood ratio is inconsistent and should be avoided (Amorim, 2008).
If calculations are performed for each type of profiling, details of both
analyses should be conveyed in the forensic report. If inclusions drawn
for both profile types support the hypothesis that a certain donor con-
veyed the DNA, the support cannot be quantified.
References
Amorim, A. 2008. A cautionary note on the evaluation of genetic evidence from
uniparentally transmitted markers. Forensic Science International: Genetics, 2,
376378.
Butler, J.M. 2006. Genetics and genomics of core short tandem repeat loci used in
human identity testing. Journal of Forensic Sciences, 51, 253265.
Dettlaff-Kakol, A. and Pawlowski, R. 2002. The first Polish DNA manhunt: An
application of Y-chromosome STRs. International Journal of Legal Medicine,
116, 289291.
Foster, E.A., Jobling, M.A., Taylor, P.G. et al. 1998. Jefferson fathered slaves last
child. Nature, 396, 2728.
Gill, P., Brenner, C., Brinkmann, B. et al. 2001. DNA commission of the International
Society of Forensic Genetics: Recommendations on forensic analysis using
Y-chromosome STRs. International Journal of Legal Medicine, 114, 305309.
Honda, K., Roewer, L., and de Knijff, P. 1999. DNA typing from 25-year-old vagi-
nal swabs using Y-chromosomal STR polymorphisms in a retrial request
case. Journal of Forensic Sciences, 44, 868872.
Innocence Project. www.innocenceproject.org
Jobling, M.A. and Gill, P. 2004. Encoded evidence: DNA in forensic analysis. Nature
Review: Genetics, 5, 742751.
Mayntz-Press, K.A., Sims, L.M., Hall, A. et al. 2008. Y-STR profiling in extended
interval ( 3 days) postcoital cervicovaginal samples. Journal of Forensic
Sciences, 53, 342348.
Roewer, L. 2009. Y-chromosome STR typing in forensic casework. Forensic Science
Medicine Pathology, 5, 7784.
Roewer, L., Arnemann, J., Spurr, N.K. et al. 1992. Simple repeat sequences on the
human Y chromosome are equally polymorphic as their autosomal counter-
parts. Human Genetics, 89, 389394.
Roewer, L. and Epplen, J.T. 1992. Rapid and sensitive typing of forensic stains
using PCR amplification of polymorphic simple repeat sequences in case
work. Forensic Science International, 53, 163171.
Roewer, L., Croucher, P.J.P., Willuweit, S. et al. 2005. International Forensic
Y Chromosome User Group: Recent historical events in the European
Y-chromosomal STR haplotype distribution. Human Genetics, 116, 279289.
Roewer, L., Kayser, M., de Knijff, P. et al. 2000. A new method for the evaluation
of matches in non-recombining genomes: Application to Y-chromosomal
short tandem repeat (STR) haplotypes in European males. Forensic Science
International, 114, 3143.
SWGDAM (Scientific Working Group on DNA Analysis Methods). 2009.
Y-
chromosome short tandem repeat ( Y-
STR) interpretation guidelines.
Forensic Science Communications, 11.
Willuweit, S. and L. Roewer, L. 2007. Y-chromosome haplotype reference database
(YHRD) update. Forensic Science International: Genetics, 1, 8387.
Wolf, A., Caliebe, A., Junge, O. et al. 2005. Forensic interpretation of Y-chromosomal
DNA mixtures. Forensic Science International, 152, 209213.
chapter seven
Other DNA techniques

including mitochondrial DNA
7.1Introduction
This chapter describes DNA techniques other than autosomal DNA pro-
filing (Chapters 1 to 5) and Y-STR profiling (Chapter6) used in criminal
cases. Mitochondrial DNA is inherited maternally in the form of haplo-
types and statistics are derived in a similar fashion to paternally inherited
Y-STR profiling. New and innovative techniques continue to be imple-
mented in criminal cases along with combinations of mitochondrial and
Y-STR profiling. These techniques are often used as last resorts when
autosomal DNA profiling is unsuccessful but are still very useful.
A discussion below of the DNA analysis of bones from ancient and
recently deceased humans and other species illustrates how DNA systems
other than autosomal profiling can be used for identification purposes.
7.2DNA analysis of bone

DNA analysis of bones has been implemented forensically since work in
1991 demonstrated that DNA could be extracted from a corpse submerged
underwater for 18 months and from bone marrow from the mummified
corpse of an 11-year-old child (Hochmeister et al., 1991).
An enduring mystery from the twentieth century was the fate of
the last tsar of Russia, Nicholas Romanov II, and his family (Figure7.1).
The tsar abdicated during the Russian Revolution in 1917 and he, his wife,
and five children were exiled to the city of Yekaterinburg. According to
historical reports, all family members and their staff were executed by a
firing squad in July 1918.
A large mass grave was discovered in 1991 and DNA testing con-
firmed the identities of Nicholas, the tsarina, and three of their daughters
in the grave (Gill et al., 1994). Nuclear DNA testing of five STR markers
confirmed the sexes of the skeletons and established a familial relation-
ship. Previous mitochondrial DNA testing confirmed an ancestral rela-
tionship through the maternal line between the Duke of Edinburgh
117
Figure 7.1 A 1914 photograph of the last Russian royal family. Seated from left
to right are Grand Duchess Olga, Tsar Nicholas II, Grand Duchess Anastasia,
Tsarevich Alexei, and Grand Duchess Tatiana. Standing from left to right
are Grand Duchess Maria and Tsarina Alexandra. (Source: Harris and Ewing
Collection, U.S. Library of Congress.)
(Prince Philip) and the tsarina and her three daughters that were found in
the mass grave. Doubts persisted that these remains were in fact those of
the Romanov family because the remains of the other two children (a boy
and a girl) were missing.
In 2007, human remains were discovered by amateur archaeologists
in a small grave near the grave described above. A variety of DNA tech-
niques linked fragments of bone and teeth from the small grave with
bones from the large grave (Coble et al., 2009; Rogaev et al., 2009). The
remains were badly damaged by fire and possibly sulfuric acid. Reference
samples were provided by living relatives.
The researchers were able to obtain complete mitochondrial DNA
sequences from the charred bone fragments. Another link came through
Y-STR profiling that allowed a comparison of the Y chromosome markers
(from the paternal lineage) of the purported male heir Tsarevich Alexei
with the markers of the tsar and a number of living male descendants.
Bloodstains from a shirt of the tsar found in storage at the Hermitage
Museum in St. Petersburg* yielded a full autosomal STR profile and a
Y-STR profile that matched the putative remains found in the grave in
Yekaterinburg. The shirt was obviously stored in an environment that
* The tsars shirt was stored after he survived an assassination attempt in Japan in 1891.
Chapter seven: Other DNA techniques including mitochondrial DNA 119
did not degrade the DNA from the blood, allowing a DNA profile to be
developed more than 100 years after the blood was deposited.
A debate over whether the remains of Anastasia or Maria were in
the second grave could not be settled based on the study results (Coble
et al., 2009). Over the years, many women have claimed to be Anastasia,
including one named Anna Anderson. The results of mitochondrial
DNA analysis of 20-year-old paraffin wax-embedded samples and hair
from Anderson ruled her out as a daughter of the tsar and tsarina (Gill
et al., 1995).
A number of mass graves dating from World War I were discovered
recently in Fromelles in northern France. Some Australian and British sol-
diers killed in 1916 were buried behind what were then German lines. The
graves were excavated in 2009 and 250 remains removed. LGC Forensics in
England is performing Y-STR analysis and mitochondrial DNA sequenc-
ing mainly on teeth but also bones; 75 bodies have been identified to date
(Thomson, 2010).
However, burnt, charred, and otherwise damaged bones present chal-
lenges in obtaining sufficient DNA for analysis. If DNA is exposed to fire
or natural elements for any length of time, degradation can occur. A loss
of signal may result from the presence of inhibitors and/or the DNA is too
fragmented to analyze. Thus careful optimization of all of the stages in
the procedure of the analysis is mandatory.
A study (Fondevilla, 2008) examined a charred femur from a major
forest fire. Mini-STR profiling and SNP (single nucleotide polymorphism)
techniques were used to confirm identity by comparison with the alleged
daughter of the male deceased. Mitochondrial DNA and Y-STR profiling
would not have been useful because mitochondrial DNA is carried mater-
nally and Y -STRs are carried paternally.
A more recent study has shown that the efficacy of obtaining DNA
from burnt bones depends on the extent of burning (Schwark et al.,
2011). Reliable DNA results could be obtained from well-preserved and
semi-burnt bones. The DNA of burnt black bones was highly degraded
and often no nuclear DNA was left, leaving mitochondrial DNA as an
option. Bluegray burnt bones yielded sporadic results and bluegray
white bones barely produced reliable results.
7.3Mitochondrial DNA basics

In addition to containing nuclear DNA, cell nuclei surround structures
called mitochondria that essentially function like power plants by pro-
viding tools for cells to make energy. The mitochondria are about the size
of bacteria and are scattered outside a cell nucleus. The mitochondrial
genome is only 1/200,000 the size of the nuclear genome, occurs in many
(about 500 to 1,000) copies per cell, and is maternally inherited without
recombination. Because mitochondrial DNA (mtDNA) is inherited mater-
nally, brothers and sisters will have the same mtDNA as their mother,
maternal aunts and uncles, and maternal grandmother. Mitochondrial
results are much less discriminatory than those from autosomal DNA
profiling.
Mitochondrial DNA shares many of the theoretical disadvantages of
Y-STR profiling that were discussed in Chapter6. It is n onrecombining so
that markers do not segregate independently, thus reducing diversity. It is
uniparentally inherited through the mother so that members of the matri-
linear line share the same haplotype. Mitochondrial DNA shows marked
population substructure and presents the complication of heteroplasmy
(see below).
It has been accepted that Y-STRs are easier to analyze than mitochon-
drial DNA. There are more haplotypes for Y-STRs and they have larger
population databases. Y -STR typing is performed at 12 or 17 loci in a sin-
gle multiplex PCR assay, compared to mitochondrial sequence analysis
across at least 610 nucleotides (and multiple strands) and often several
amplifications with difficult samples.
The major advantage of mitochondrial DNA is its multiple copy num-
ber per cell. This means that it has a greater probability of survival than
nuclear DNA. Forensic applications include analysis of old, degraded
and/or damaged samples and samples such as hair shafts that contain
low levels of nuclear DNA.
Case 1 was the first U.S. criminal proceeding that introduced mito-
chondrial DNA profiling results from hair as trial evidence (Davis, 1998).
Case 1
A Tennessee murder trial in 1996 convicted Paul Ware of the
rape and murder of a 4-year-old girl. The defense claimed that
the babysitter framed Ware, who was found drunk and asleep
next to the body of the child. His semen was not found on the
child. However, during the autopsy, a short red hair was found
in the throat of the child and several red hairs were found in
the bed at the crime scene.
Mitochondrial DNA was extracted from two of the hairs
one from the throat of the victim and one from the bed where
the offense was believed to have occurred. The mitochondrial
DNA of the hairs was compared to and found to match Wares
mitochondrial DNA. The haplotype had not been observed in
an FBI database of 742 individuals.
Case 2 (Innocence Project) further shows the advantages of mito-

chondrial DNA profiling on hair shafts. Evidence believed to have been
destroyed was later located and was successful in clearing a wrongfully
accused male.
Case 2
In 1995, two men carrying pistols and wearing ski masks burst
into an apartment in Oklahoma and bashed and robbed a young
woman. She identified Sedrick Courtney as wearing a black
ski mask (he lifted the mask before he bashed her). Black and
green ski masks were discarded outside the victims apart-
ment. Courtney was arrested despite having an alibi. The other
robber was never found.
Hairs were found in both ski masks and the roots were
sent for autosomal DNA profiling; no results were gener-
ated. A forensic analyst stated that the hairs in the black mask
could not be eliminated as having come from Courtney and a
bleached red hair in the green mask was microscopically con-
sistent with a bleached hair from Courtney. The prosecution
stated the accused could have owned both masks and hairs
from both masks could have come from him.
In 2000 and 2007, Courtney requested DNA testing on
the hairs but the police said the hairs had been destroyed.
Courtney was paroled in 2011 and a lawyer for the Innocence
Project learned that the police retained the hairs on a micro-
scope slide. Mitochondrial DNA excluded Courtney from the
hairs on the black and green masks and the charges against
him were dismissed in July 2012.
Mitochondrial DNA is defined by current convention into three regions: a

coding region and hypervariable regions I and II. The DNA sequences of
the hypervariable regions are usually analyzed as they contain the largest
proportion of diversity in the mitochondrial genome.
Heteroplasmy can cause different mitochondrial DNA sequences to be
found in different tissues from a single individual, even along the length of a
single hair shaft. Mutation, which distinguishes heteroplasmic types, is par-
ticularly common at some sites (hot spots). Shared heteroplasmy between
two samples can actually increase the strength of evidence. This was shown
in the confirming of the matrilineal relationship between the remains of
Tsar Nicholas II and his brother, Georgii Romanov (Coble et al., 2009).
Case 3 is from a high profile 2004 California murder trial that involved
a mitochondrial DNA match.
Case 3
Scott Peterson was convicted in 2004 of murdering his wife
(Laci) who was eight months pregnant with their child. The
fetus was found washed up in San Francisco Bay in April 2003
and Lacis torso was located the next day. The exact date and
cause of death were unknown. Peterson reported his wife
missing on Christmas Eve 2002.
The only piece of forensic evidence was a six-inch black hair
wrapped around a pair of pliers on Scotts boat. His wife had
never been on the boat. Mitochondrial DNA evidence from the
hair was shown by the prosecution to be consistent with Lacis
mitochondrial DNA; her mother provided the reference sample.
The defendant had a different mitochondrial haplotype.
The haplotype of the hair from the pliers was expected once
in every 112 Caucasians, from a database of 1,833 individuals.
Defense lawyers challenged the evidence as unreliable.
First, they declared that mitochondrial DNA was novel and
not generally accepted in the scientific community. Second, the
statistical probability in the case was insignificant and ambig-
uous and therefore incapable of helping the finders of fact in
the dispute (Geragos, 2003). Alternatively, if the mitochondrial
DNA met the standards, the defense protested that the careless
actions of the police exposed the items to significant risks of
alteration and contamination. The judge admitted the evidence
(Girolami, 2003).
Peterson has always maintained his innocence and is on death row; an

appeal was filed in July 2012 (Elian, 2012).
7.4Statistics in mitochondrial DNA analysis

The first stage in assessing the value of a mitochondrial DNA profile is to
consider the relevant population from whom the biological material may
have originated. Similar to Y-STR profiles, and because haplotypes from a
database are compared, the relevant population depends on geographical
factors and other circumstantial information (if any) known at the time.
Could the donor of the material be a suspect in the murder of his business
associate? Could a donor be the victim of a plane crash?
The databases used may be different. In the absence of case-specific

information, international databases of mitochondrial DNA profiles
that are available in print and electronically may be consulted to try to
infer a donors ethnic appearance. The widely used EMPOP 9 database
(www.empop.org) contains the profiles of about 4,500 individuals, mostly
of European origin.
The usual method of reporting the significance of matching profiles
is to state the number of times (relative frequency) that this profile has
been seen in the relevant database. For example, in the context of criminal
proceedings in which Mr.X is the suspected donor of a hair with a certain
mitochondrial DNA haplotype, an expert might state:
In my opinion, the mitochondrial DNA sequence

from Mr.X, which matches that of the crime stain,
would be expected to be observed in fewer than 1
in 2,000 randomly chosen people in the European
Caucasian population.
A High Court of Australia appeal concerned the provision of two

types of statistics in a murder trial (Ayturgrul v. The Queen, 2012, also dis-
cussed in Chapter4). It was agreed in the trial that the appellant could
have been the donor of a hair found on the deceaseds thumbnail. The
statistics of a matching mitochondrial DNA profile were given as: (1) 1 in
1,600 people in the general population would be expected to share the
same haplotype (frequency ratio); and (2) 99.9% of people would not be
expected to have a haplotype matching the hair (exclusion percentage).
The appellant asked to have the exclusion percentage figure judged
inadmissible. He held that this percentage figure produced a subliminal
or subconscious impact that invited the jury to approach the case with
percentages of guilt and round the figure up to 100%.
The prosecution expert used the counting method of observing the
haplotype one time in a database of 4,839 individuals of various popula-
tion groups. A defense expert was of the opinion that 1 in 1,000 people in
the non-Turkish population would have this haplotype and 1 in 50 people
in the Turkish population would have the haplotype (the appellant was of
Turkish descent). These appear to be very different statistics but they are
dependent on their derivation.
During the original trial, the method of deriving the statistics and the
fact that mitochondrial DNA profiling was much less discriminatory than
nuclear DNA profiling were discussed. The High Court dismissed the
appeal as it did not consider that the appellant demonstrated that the pro-
bative value was outweighed by the danger of unfair prejudice.
7.5Contamination
A major concern in mitochondrial DNA analysis is contamination of a
sample with extraneous DNA (Isenberg, 2005). Mitochondrial DNA anal-
ysis is a very sensitive technique and the presence of low level contamina-
tion is not uncommon.
Contaminant DNA from any source can appear as a major or minor
sample within a mixture or may overwhelm the target DNA completely.
Gross or sporadic contamination can appear before an incident has
occurred, in the interval between the incident and securing the scene,
during the investigation of the scene, and/or within the laboratory.
Mixed mitochondrial DNA profiles in a bone sample indicate that
contamination (extraneous DNA not inherent to the bone) has been intro-
duced into the sample. A single bone or bone fragment belonging to a
human should not have a mixed mitochondrial DNA profile that indi-
cates contributors from at least two individuals. Case 4 from the authors
files was a homicide involving mixed mitochondrial DNA profiles in
bone samples.
Case 4
A man was reported missing in Melbourne, Australia. Bone frag-
ments were found at a beach location about a two-hour drive away.
It was alleged that the accused murdered the victim, burned his
body in a drum, and disposed of the bones at the beach.
One bone fragment failed to yield a nuclear DNA profile
but gave a mixed mitochondrial DNA profile suggesting the
presence of at least two mitochondrial DNA profiles. The pros-
ecution expert stated that the bones were human due to the
presence of mitochondrial DNA. Mitochondrial DNA from the
bone powder was amplified and sequenced by another labo-
ratory in the United States. Its sequence data also indicated a
mixture of two or more mitochondrial DNA profiles and there-
fore the results were inconclusive. Because of the mixture pro-
file of mitochondrial DNA and inability to assign a profile to
the bone, the bone could not be determined to be human.
If the bones were not originally cleaned sufficiently or a
contaminant was introduced at the first examining labora-
tory, these errors will necessarily impact subsequent testing by
another laboratory. However, if the cleaning techniques before
the pulverization of the bones into powder met the standards,
any touch DNA acquired through handling before introduc-
tion into the laboratory that remains on the surfaces of the
bones has the possibility to be removed or considered. This is

another reason to provide another analyzing laboratory with a
whole item so that its contamination mitigation measures can
be utilized. The mitochondrial DNA evidence was not admit-
ted into trial. The accused was found guilty on other evidence.
The ability of PCR (see Chapter3) to amplify small amounts means that
when ancient specimens contain little or no endogenous DNA, the DNA
amplified may be derived partially or wholly from exogenous DNA con-
tamination and be mistaken for endogenous (inherent) DNA. As an example,
a report of dinosaur DNA sequences actually proved to be derived from
human DNA contaminating the fossil (Malmstrom, 2005). All samples
from 29 dog museum archaeological specimens contained human DNA
often at levels exceeding authentic ancient dog DNA.
Bones and teeth are normally the longest lasting physical evidence
of human or animal presence and are also the most widely used sam-
ples for ancient DNA studies. However, they are readily contaminated
(presumably through handling and washing) and difficult or impossible
to decontaminate after such contamination. Sparse, damaged endogenous
DNA is less likely to be amplified than modern contamination. Current
techniques used to decontaminate specimens include the application of
bleach, exposure to UV light, and grinding or shot blasting, and reflect a
belief that contamination is concentrated in the outer surface of a mate-
rial. Even when strict protocols are followed contaminants are frequently
observed. Human DNA has been reported from cave bear samples,
500-year-old pig samples, and 109 of 168 relatively recent fox teeth. Several
studies have reported significant numbers of human remains contami-
nated with multiple human sequences (Gilbert et al., 2005).
Knowledge of the history of the sample handling prior to the analysis
is thus critical. The important early Australian Mungo man study (Adcock
et al., 2001) did not mention that the sample had been excavated in the
1970s and handled many times after that. Thus, it was likely to be contam-
inated, contained negligible organic preservation, and was considered too
fragmentary to sex reliably because of very poor preservation practices.
Without such information, it is very difficult to comment objectively on
the reliability of results (Gilbert et al., 2005).
7.6Mixture mitochondrial DNA profiles

Unlike results from standard DNA and Y-STR profiles, it is not currently
possible to determine the number of donors or their respective contribu-
tions to a mixed sample using mitochondrial DNA. Some progress has
been made in separating mixed mitochondrial DNA profiles (Gilbert et al.,

2005) but the techniques are not yet operational in forensic laboratories.
7.7Familial DNA searching

DNA databases are designed to provide law enforcement with investi-
gative leads and rely on full or partial DNA matches (all alleles avail-
able) with crime scene samples. Familial DNA searching utilizes partial
matches (not partial profiles) and further DNA interpretation is required.
Close family relatives are used, for example a father and son who share
one allele at each locus according to Mendelian inheritance (see Chapter1).
The United Kingdom pioneered the use of familial DNA searching.
Case 5 was one of the first criminal matters utilizing this technique to
solve the Valentines Day murder of Lynette White in 1988 in South Wales.
Case 5
The victim was murdered in her bedroom. Investigators found
more than 50 stab wounds, an almost severed head, and indi-
cations of male blood at the scene. Her flat was above a bet-
ting shop in Cardiffs red light district. Five local men, one of
whom was her pimp, were arrested.
The first trial did not finish due to the death of the judge.
The second trial was held in 1990 and three of the five accused
men were convicted of murder. However, the convictions were
quashed on appeal in 1992.
A cold case review was launched in 2000 and all the
exhibits were sent to a different laboratory, Forensic Alliance
(Exhibit A, 2004). The crime scene yielded 954 exhibits and a
partial male DNA profile was obtained from a blood spot on a
cellophane cigarette packet. It was postulated that one offender
may have cut himself during the frenzied attack.
The scientists went back to the original flat and found that
it had been repainted. However they examined the skirting
board below the original splashes of blood seen in the crime
scene photographs. Three weeks of scraping back the paint
revealed traces of the original bloodstains, and a full DNA
profile was obtained from the blood that matched cellophane
man. Eventually his blood was found in 10 places in the flat,
on and around the body, and along the exit route.
The DNA profile was placed on the DNA database but no
match was found. However, the profile exhibited an uncom-
mon allele, so the profile was searched on the South Wales
police submissions database for a family type match. One pro-

file that stood out was from a 14-year-old boy born after the
murder. His uncle, Jeffrey Gafoor, gave a DNA sample to police
and it matched cellophane man. He pleaded guilty and was
convicted in 2003.
When a routine search of a DNA database does not show that a crime
profile matches any profile in the database, it is possible to conduct a
search to identify potential relatives of the donor of a crime stain. This
search is based on the number of shared genetic characteristics (alleles)
and the rarity of the shared alleles in human populations.
Unlike a search for a direct match, a familial search will allow for
matching subsets of alleles at any given genetic marker as a basis for com-
parison. A familial search relies on mathematical modeling specific to the
DNA database utilized. This modeling determines whether an observed
similarity between two DNA profiles is more likely the result of kinship
or mere chance (Myers et al., 2010). This analysis is more time consuming
and labor intensive than traditional nuclear DNA testing.
Not all jurisdictions have supported the technique and concern sur-
rounds the value of familial DNA testing when balanced against privacy
issues. Case 6 (Butler, 2012; Steinhauer, 2010) is a famous success story,
although the trial has still not been held. At the time of the arrest of the
accused, only two states (California and Colorado) had codified policies
permitting familial searches. The Grim Sleeper case was the first use of
an active familial search to solve a homicide in the United States.
Case 6
Lonnie Franklin Jr. was arrested in July 2010 as a result of a
familial DNA search in California. He is currently accused of the
murders of 10 young women from 1985 to 2007. The women were
murdered in Los Angeles and the cases were linked through
firearms analysis and DNA. The perpetrator was called the
Grim Sleeper because of the 13-year gap in detected crimes.
A familial search of DNA database profiles in 2010 yielded
one likely suspect based on the crime scene profiles from a
DNA profile that was added to the database in 2009 after a
felony weapons charge. Profiles from the Grim Sleeper crime
scenes shared 1 allele at all 15 loci with the felon on the weap-
ons charge. This meant that it was possible that the felon was
a relative of the Grim Sleeper. They also shared the same
Y-STR profile.
The Los Angeles police had a suspect, the father of the

felon, and followed him. They assigned an undercover police
officer to act as his waiter in a pizza restaurant and the waiter
collected his discarded utensils and pizza leftovers so that a
comparison DNA profile could be generated. The result was
a full nuclear DNA match with the crime scene profiles. The
accused pleaded not guilty.
The technique still remains controversial due to privacy concerns and

technical pitfalls (Butler, 2012), but there is no doubt it can be invaluable
when no investigative leads are generated.
7.8Domestic animal hair

Many households in every country keep domestic pets such as cats and
dogs. These animals readily shed hair on the clothing and environment of
humans (except for certain nonshedding breeds such as poodles). It is thus
likely that this type of evidence can appear at crime scenes on the clothing
of the victim and/or perpetrator or on items like furniture. Today, the abil-
ity to profile DNA of animal hairs has increased the potential of this type of
evidence. However, in general case work experience, this type of evidence
is ignored or used as a last resort. It is worthwhile for legal professionals to
recognize that animal hair may be used when other avenues have failed.
Case 7 was the first criminal matter involving animal hair DNA. The
hair came from Snowball, a white cat belonging to the parents of a murder
suspect (Menotti-Raymond et al., 1997).
Case 7
The body of 32-year-old Shirley Duguay was found in a shallow
grave in a wooded area of Prince Edward Island in Canada in
1994, some eight months after she disappeared. A mans jacket had
been found 8 km from her house three weeks after she had gone
missing. The leather jacket was covered in bloodstains matching
the deceased and many white cat hairs were found on the lining.
Douglas Beamish, the victims estranged boyfriend, lived
with his parents who had a white cat called Snowball. The DNA
profile of the cat hair on the jacket matched that of Snowball
(Menotti-Raymond et al. 1997; Coyle, 2008).
The case set a legal precedent allowing animal DNA to be
admitted as evidence in criminal trials. Beamish was convicted
of murder and sentenced to 15 years in prison.
Cats have 18 pairs of autosomes and the X and Y sex chromosomes. The
commercially available MeowPlex kit contains 11 STR markers (Butler
et al., 2002). Dogs have 38 pairs of autosomes as well as sex chromosomes.
7.9Other techniques
A Minifiler STR kit uses STR markers that are reduced in size compared
to standard STR kits used for routine DNA profiling (e.g., ProfilerPlus in
Australia and the AmpFLSTR Identifiler in the U.S.). The reduced sized
amplicons enable higher recovery of information from degraded DNA
samples by improving amplification efficiency (Butler, 2012).
DNA is now being applied to botany in cases where plant material
from a scene may be linked to a suspect or victim. Microbial forensics is
an emerging field that studies variations in bacteria and viruses. These
new techniques must be shown to be valid and relevant before they are
accepted as scientific evidence in courts of law.
References
Adcock, G., Dennis, E., Easteal, S. et al. 2001. Mitochondrial DNA sequences in
ancient Australians: Implications for modern human origins. Proceedings of
the National Academy of Science of the USA, 98, 537542.
Ayturgrul v. The Queen. 2012. High Court of Australia, HCA 15. Respondents sub-
missions to the High Court, S315, filed October 21, 2011.
Butler, J.M. 2012. Advanced Topics in Forensic DNA Typing: Methodology. San Diego:
Butler, J.M., David, V.A., and M enotti-Raymond, M. 2002. MeowPlex: A new DNA
test using tetranucleotide STR markers for the domestic cat. Profiles in DNA,
710.
Coble, M.D., Loreille, O.M., Wadhams, M.J. et al. 2009. Mystery solved: The identi-
fication of the two missing Romanov children using DNA analysis. PloSONE,
4, March 11.
Coyle, H.M., Ed. 2008. Nonhuman DNA Typing: Theory and Casework Applications.
Boca Raton, FL: CRC Press.
Davis, C.L. 1998. Mitochondrial DNA: State of Tennessee v. Paul Ware. Case Report.
Profiles in DNA, GenePrint. www.promega.com
Elian, P. 2012. Scott Peterson appeals death sentence for 2002 murder of wife Laci
Peterson. Huffington Post, July 6.
Exhibit A. 2004. News from Forensic Alliance, 1.
Fondevilla, M., Phillips, C., Naveran, N. et al. 2008. Case report: Identification of
skeletal remains using short amplicon marker analysis of severely degraded
DNA extracted from a decomposed and charred femur. Forensic Science
Geragos, G. (Attorney for Defendant Scott Peterson). 2003. Notice of motion and
motion in limine to exclude mitochondrial DNA evidence. People of the State
of California v. Scott Lee Peterson, Superior Court Case 1056770. October 20.
Gilbert, M.T.P., Rudbeck, L., Willerslev, E. et al. 2005. Biochemical and physical
correlates of DNA contamination in archaeological human bones and teeth
excavated at Matera, Italy. Journal of Archaeological Science, 785793.
Gill, P., Ivanov, P.L., Kimpton, C. et al. 1994. Identification of the remains of the
Romanov family by DNA analysis. Nature Genetics, 6, 130136.
Gill, P., Kimpton, C., Aliston-Greiner, R. et al. 1995. Establishing the identity of
Anna Anderson Manahan. Nature Genetics, 9, 910.
Girolami, A. (Superior Court Judge). 2003. Mitochondrial DNA evidentiary rul-
ing. People of the State of California v. Scott Lee Peterson, Superior Court Case
1056770. November 18.
Hochmeister, M.N., Budowle, B., Borer, U.V. et al. 1991. Typing of deoxyribonucleic
acid (DNA) extracted from compact bone from human remains. Journal of
Isenberg, A. 2005. Forensic mitochondrial DNA analysis. In Saferstein, J., Ed.
Forensic Science Handbook, Vol. II. New York: Pearson Prentice Hall.
Malmstrom, H., Storra, J., Holmgund, G. et al. 2005. Extensive human DNA con-
tamination in extracts from ancient dog bones and teeth. Molecular Biology
and Evolution, 22, 20402047.
Menotti-Raymond, M., David, V.A., and OBrien, S.J. 1997. Pet cat hair implicates
murder suspect. Nature, 386, 774.
Myers, S.P., Timken, M.L., Piucci, G.A. et al. 2010. Searching for fi rst-degree famil-
ial relationships in Californias offender DNA database: Validation of a likeli-
hood ratio-based approach. Forensic Science International: Genetics, November.
Rogaev, E., Grigorenko, A., Moliaka, Y. et al. 2009. Genomic identification in the
historical case of the Nicholas II royal family. Proceedings of the National
Academy of Sciences of the USA, 106.
Schwark, T., Heinrich, A., Preusse-Prange, A. et al. 2011. Reliable genetic identi-
fication of burnt human remains. Forensic Science International: Genetics, 5,
393399.
Steinhauer, J. 2010. Grim Sleeper arrest fans debate on DNA use. New York Times,
July 8.
Thomson, J.A. 2010. The Fromelles War Graves Project: Identification of skeletal
remains from World War I. Haploid DNA markers in Forensic Genetics
Conference, Berlin. http://www.lgc.co.uk
chapter eight
Concerns and controversies

This chapter will discuss situations in which DNA results may be com-
promised through the scientific processes that yielded them or through
the interpretation of their meanings in the context of a case. Issues, such
as contamination and error rates that affect the quality and subjectiv-
ity, and transfer that affects the reliability and validity of the results will
be covered. A roll call of true but bizarre cases from England, Australia,
the United States, and Europe will illustrate these points.
At the end of this chapter, the obligations of the reporting scientist
and the criminal justice professional involved in a case will be discussed.
8.1Introduction
Evidence can arise (1) through innocent means, (2) as a result of the crime,
and (3) as a result of contamination or inadvertent transfer. The mecha-
nism of transfer of a DNA profile is a consideration for every case reported.
Two well known authors in the field emphasize the responsibility for a
scientist to place the evidence in context and point out the limitations of
interpretations (Gill and Buckleton, 2010). Limitations of the DNA results
should be conveyed clearly in both the report and testimony of the foren-
sic scientist.
The legal practitioner should be alert to a few warning signs when
examining a case involving DNA evidence. Cold cases are often
re-examined, but the exhibits may have been initially examined in con-
ditions lacking the strict contamination mitigation measures used today.
This is because transfer of minute quantities of DNA was not considered
before the 1990s. Another warning sign is a single exhibit producing DNA
results. The more DNA results of evidential value generated supporting
the prosecution hypothesis, the less likely contamination, transcription,
or other errors may have taken place.
This is especially true if several body fluids or t wo-way transfers are
involved. Consider a case in which a DNA profile from blood found on
the suspects clothing matches the victim and DNA from semen found
on the victim matches the DNA of the suspect. Contamination would be
less likely in this case than in a rape case in which DNA on a single medi-
cal swab from the victim matches the profile from the accused.
131
Finally, could the DNA have been deposited through legitimate con-
tact based on the principles of trace evidence transfer? The method of
deposition of the DNA should always be considered and various path-
ways proposed as a step in following the scientific method.
8.2Quality issues
Quality is considered the ability of a procedure or product to satisfy a need,
be free of defects, and meet a set of requirements. Quality in a forensic sci-
ence laboratory is controlled by a quality assurance system and should
be strictly monitored, especially if the laboratory is accredited. A number
of procedures may be followed if the quality of forensic evidence is sus-
pected to be an issue in a criminal or civil case. These include audit trails
of samples, validation studies of the test involved, peer reviews, internal
and external audits, proficiency testing, and maintaining expertise.
The processing of a DNA sample requires many steps and each step
may present the potential for error. These steps include identifying the
biological stain or material on the item, extracting the DNA, quantifying it,
amplifying it, separating the components, and finally interpretation includ-
ing statistical interpretation. Many cases are usually processed simultane-
ously, following each procedural step. Case 1 shows how contamination
between two cases occurred in the extraction step of the DNA analysis.
Case 1
Adam Scott, a 19-year-old male from Devon in England was
accused of raping a woman in October 2011 in Manchester. He
claimed he had never been to Manchester (Morris, 2012).
Scott subsequently spent five months in jail on remand in
custody after a database search allegedly found that his DNA
profile matched that from semen found on a vulval swab of the
victim. He was released in March 2012 after being found the
innocent victim of an avoidable contamination (Rennison, 2012).
Scotts saliva from an unconnected earlier case preceded
the pertinent medical swab sample taken from the alleged rape
victim; it wasin an earlier batch of samples through the extrac-
tion process. The DNA material from the victims medical swab
was the sole evidence against the accused.
DNA profiles from the seminal fractions of the two low
vaginal swabs, high vaginal swabs, and a vulval swab matched
the victims boyfriend. A second vulval swab produced a mixed
profile from the victims boyfriend and an unknown male
(17 of 20 alleles present). The unknown male DNA profile was
Chapter eight: Concerns and controversies 133
loaded onto the national DNA database. The search yielded a

partial DNA profile match of Scotts DNA with a probability of
one in one billion.
The investigation revealed that a plastic sample holder tray
that should have been discarded was mistakenly re-used and
loaded into equipment by a laboratory worker to undergo robotic
extraction. Scotts saliva sample had been processed earlier in the
re-used tray andin the same well (sample container slot)that
was used later for the vulval medical swab from the alleged rape
victim. It wasdetermined thatthere was sufficient DNA present
in the well from the earlier case to contaminate the later alleged
rape sample.
Guidelines in place in various jurisdictions around the world recom-

mend that suspects should not be prosecuted on DNA evidence alone.
These guidelines should serve as safeguards but they were not observed
in Scotts case.
Continuity and audit trails may assist in querying the potential for
error at each step. An audit trail of a sample should have a unique identi-
fier that enables the sample to be followed at each stage of the analysis.
This allows questions such as which scientist handled the sample on a
particular date and location to be answered readily.
The cases of O.J. Simpson (Case 2, Chapter1) and the Omagh bombing
(Case 1, Chapter5) showed that poor collection procedures at crime scenes
preceding exhibit arrival at a laboratory compromised both cases. Crime
scene personnel today should follow the proper procedures for collecting
exhibits and prevent the ready transfer of their own DNA to the scene and
any exhibits they handle.
8.3Relevant sample testing

Every forensic report should incorporate the scientific method that
includes a proposal of differing hypotheses and a testing rationale. A
perusal of the case notes should elicit this information but if it cannot be
deciphered, the legal professional should request clarification from the
case-reporting scientist.
Sometimes the relevant samples in a case are not tested. The bedding
in the Frank Button case (Case 8, Chapter2) was not tested initially. The
defense later requested testing of the bedding and the result revealed
semen with a DNA profile different from Buttons. The previously unsuc-
cessful DNA result from medical swabs from the rape victim was also
r e-analyzed and exhibited a profile matching that of a convicted rapist.

Button was acquitted.
8.4Contamination
Contamination is an issue that may contribute to an error in a forensic
result. It may result from a quality failure, for example, improper or careless
handling introduces extraneous substances into a sample. Contaminants
can be introduced at any stage of the testing process, from crime scene
collection to the final result. Contamination of evidence has contributed to
miscarriages of justice in many prominent cases. Cases 2 and 3 illustrate
contamination of clothing exhibits.
Case 2
The high profile murder investigation of the death of Jaidyn
Leskie in Victoria, Australia involved laboratory contamination
of evidence samples from different cases. The toddlers body
was found in a dam near Moe in 1998, some six months after
he went missing. A trial jury acquitted the mothers ex-partner
and babysitter, Greg Domaszewicz, of the murder in 1998.
A DNA profile was obtained in 2003 from the childs clothing
found in the dam; it matched the DNA profile obtained from a
condom taken as evidence in a rape case. The police could find no
connection between the rape victim and the murdered toddler.
The inquest (Johnson, 2006) discovered that the childs
clothing was examined within days of the examination of the
condom from the rape case by the same forensic scientist in
1998. The coroner found that contamination occurred in the lab-
oratory although the exact pathway could not be determined.
Another example of contamination between evidence samples occurred

in the same Victoria laboratory during the cold case investigation of a
double murder (Hadfield, 2011).
Case 3
Margaret Tapp and her young daughter, Seana, were mur-
dered in 1984 in Victoria. In 2008, a DNA database hit matched
a profile from Seanas nightwear with a reference sample from
Russell Gesah.
Detectives could find no link between Gesah and the Tapps
and it was discovered that he was not in the state of Victoria
at the time of the murders. An inquiry found that in 1999 an

unrelated exhibit from Gesah was tested on the same day at
the same laboratory as exhibits from the murder scene. Gesah
was released from custody two weeks after his arrest; the mur-
ders are still unsolved.
A report to the Victorian Parliament concerned the 2008 conviction of

Farah Jama for the rape of an unconscious woman in a nightclub (Vincent,
2010; also see Chapters 2 and 4). The DNA evidence was contaminated in
the examining rooms of a hospital. Justice Vincent noted in the report that
the DNA evidence was perceived to appear so powerful by all involved
in the case that none of the filters on which our criminal justice system
depends to minimize the risk of a miscarriage of justice operated effec-
tively at any stage of the case until a few weeks before Jamas appeal.
Vincent noted that no one appeared to be aware of the dangers of relying
on statistical probabilities in the determination of guilt.
The appeal court in the Meredith Kercher murder commented on
contamination of exhibits collected at crime scenes, particularly low level
DNA (Hellmann, 2011). An original statement was that it was not enough
for the defense to claim that the DNA result was from contamination. The
burden was on those claiming contamination to prove its origin. However,
the appeal court held that the burden was proving that the result was
obtained using a procedure that guaranteed the integrity of the item from
the moment of collection to the moment of analysis. If no proof demon-
strates that these precautions were followed, it is not necessary to also
prove the specific source of the contamination (Hellmann, 2011).
One way to prevent contamination is to ensure that reference and
evidentiary samples are kept separate in time and space and that separa-
tion procedures are followed routinely in forensic laboratories. Prevention
of contamination between evidentiary samples in a single case and even
between cases, as seen from the above case studies, requires extra mea-
sures. No single specific test reveals contamination. However, audit trails
may follow the path of an exhibit from collection to final analysis and
indicate areas in which contamination may be a possibility.
The laboratory environment in which DNA analyses are performed
should be rigidly controlled. Leading U.K. forensic laboratories utilize
requirements such as wearing disposable laboratory overalls, hair nets,
face masks, and protective shoes on entry and disposing of the items on
exit. Weekly monitoring of the background levels of DNA in laboratory
rooms is also performed to ensure contamination mitigation.
Poy and van Oorshot (2006) examined the levels of background DNA
in a forensic laboratory. They studied laboratory and office areas at 195
sites and categorized them according to their contamination risk levels.

Allelles were present at 52 of the sites (27%). Of 32 interpretable profiles,
28 matched staff members, mostly in the office area where no protective
clothing was required; of the 4 unknown profiles 2 matched those in the
criminal database. One profile recovered from a magnification lamp was
believed to have come from a bulky, heavily stained sample examined
3 months prior to the study. The authors stressed the need for rigid con-
tamination prevention.
A further study was undertaken in Victoria to detect DNA on unused
gloves from opened and unopened boxes in a forensic laboratory (Daniel
and van Oorshot, 2011). The study revealed the presence of DNA on gloves
from the closed boxes and on some gloves from opened boxes. It was rec-
ommended that only certified DNA-free gloves be used in the laboratory
and regular contamination monitoring of gloves be performed.
An interesting study was performed in New South Wales to answer
questions raised in two court cases. Allegations that police officers may
have inadvertently transferred suspect DNA to drug balloons and fire-
arms as a result of a search process were put forth (Beilby, 2006). The sug-
gestion was made that transfers may have occurred through handling
items belonging to the suspect (bag, clothing, other personal items), then
handling other items while wearing the same gloves. The study found
that secondary transfer to drug balloons was possible immediately after a
search of a bag belonging to a volunteer, but did not occur if other objects
were handled between the handling of the suspects items and the other
evidence. Secondary transfer did not occur with firearms. The study rec-
ommended that police change gloves regularly during searches.
Other avenues of contamination should be considered. Using collec-
tion and detection devices on multiple exhibits and at multiple scenes
can introduce minor or gross contamination. Fingerprint brushes, for
example, can transfer DNA between exhibits that could generate profiles.
The brushes may also retain biological evidence for a considerable time
(van Oorshot et al., 2010). DNA deposited on one item can thus be trans-
ferred to another.
Case 4 from Germany confounded investigators and highlights the
power of DNA evidence even when other evidence appears contradictory
(Himmelreich, 2009).
Case 4: The Phantom of Heilbronn

Police linked 40 crime scenes incorrectly. One of these cases
was the murder of a 22-year-old policewoman in the town
of Heilbronn in southern Germany attributed to a female
serial killer. DNA supposedly from the phantom (woman
without a face) was found on a wide variety of items includ-

ing a cookie, a heroin syringe, and a stolen car. In March 2009,
it was revealed that all the cotton swabs used to collect samples
from the 40 crime scenes may have been contaminated by the
same female worker in a factory in Austria and no phantom
ever existed.
The phantom gained notoriety in 2007 after the murder of
the policewoman. The police announced that they found the
phantoms DNA traces on several cold cases including a homi-
cide from 1993. However, the types of cases committed (the
phantom was both a brutal killer suspected of at least six mur-
ders and a common thief) were widely disparate. She had been
involved in a car dealership burglary and a school break-in.
In both cases, her convicted accomplices denied her existence.
The real phantom was discovered when officials trying
to establish the identity of a burned corpse from fingerprints
on an asylum application form found the form to contain the
phantoms DNA. The finding was considered impossible and
officials repeated the analysis and found no DNA. They ulti-
mately discovered that the cotton swabs used to collect DNA
material were contaminated. Although they were sterilized
for medical use, the sterilization process did not destroy DNA
and some of the swabs contained enough cellular DNA to yield
profiles. There had been a collective suspension of disbelief
about the DNA evidence even among sophisticated detectives
that trumped all other facts.
As a measure to prevent cases like the phantom, the Scientific Working

Group on DNA Analysis Methods (SWGDAM) in the U.S., the European
Network of Forensic Science Institutes (ENFSI), and the Biologist Special
Advisory Group (BSAG) in Australia have issued a position statement
with specific recommendations for manufacturers and laboratories (Gill
et al., 2010) to ensure that all materials used are D NA-free, rather than
simply sterile.
Police and crime investigators must invest greater effort in investigat-
ing and documenting how DNA samples arrived where found. Scientists
must better understand the impacts of activities on the relative amounts
of DNA from sources at a crime scene. In some instances, it is possible to
derive the chain of events that led to the presence of a DNA sample at a
crime scene, for example, prior visits to a scene or known use of an item.
Awareness of these variables and their impacts on transfer events will
help weigh the likelihoods of proposed alternative scenarios (van Oorshot

et al., 2010).
Another case from Victoria showed that significant quantities of DNA
are frequently (1) transferred from an exhibit to the inside of its packag-
ing and (2) transferred from its area of initial deposit to other areas of the
same exhibit and/or to other exhibits within the same package (Goray
et al., 2012). These findings highlight the need to deal with issues inherent
in the collection and packaging of exhibits for forensic DNA analysis.
8.5Interpretation issues
Interpretation of DNA profiles is relatively straightforward when a sam-
ple is of sufficient quantity (all peaks above the stochastic threshold, see
Chapter 5) and appears to be from a single source. It is even simpler when
the DNA can be associated to a body fluid with some confidence. If con-
firmatory tests of these biological fluids are performed, the DNA may be
associated more readily to a body material (although assumptions still
need to be made and communicated).
Problems in interpretation may occur when the DNA obtained is a
mixture of contributors (two or more people), yields only a partial profile,
is low level, or cannot be associated to a particular body fluid.
Touch DNA requires careful interpretation. The murder of Meredith
Kercher (Case 7, Chapter2) involved touch DNA on a bra clasp that gener-
ated a minor DNA profile matching a profile of one of the accused. The
result was a low level profile from a clasp collected 46 days after the crime.
The delay in procuring the clasp introduced the potential for contamina-
tion at the scene.
The appeals court also questioned why the knife in the case was
believed to be evidential. The low level DNA on the blade of the knife
found in a kitchen drawer of Sollecitos flat could not be sourced to a body
fluid. Furthermore, the victim and accused had access to each others
apartments. Why was the possible presence of the DNA of the deceased
on the blade considered evidential?
8.6Error rates
Forensic science commentators frequently ask the same question. Why is
it necessary to have match probability data on DNA profiles to determine
the rarity of a DNA profile but no data about the probability of a wrong
result? The high numbers generated in DNA profiling of match probabili-
ties appear to overwhelm all arguments.
The probability of a wrong result is a different question. How often
mistakes are made is a basic occurrence in science and is designated
the error rate, but error rates in forensic disciplines have received lit-
tle publicity.
When DNA evidence was first introduced, a number of experts testi-
fied that false positives were impossible in forensic DNA testing. As we
have seen in previous chapters, these claims are not true. Among the first
200 people exonerated by p ost-conviction DNA testing were Timothy
Durham and Josiah Sutton (Innocence Project). Both were convicted due
partly to DNA testing errors. In both cases, a combination of laboratory
technical problems and careless or mistaken interpretation of the test
results produced misleading evidence that helped send innocent men to
prison for many years.
There is little published data on error rates in forensic disciplines.
Many forensic scientists argue that it is not possible to obtain an error rate
in a specific discipline because there is no way to determine how often
erroneous results are obtained. Some scientists such as fingerprint exam-
iners claim that their discipline achieves a zero error rate. The assumption
is that no two people have the same fingerprints. However, how likely is
it that two people share a given number of fingerprint characteristics? No
available data exist to determine that likelihood.
One way to obtain error rates is to perform blind proficiency tests that
mimic real cases, although objections have been raised to blind tests on the
basis that proficiency testing can never reflect actual forensic casework.
The only study to date examining error rates in forensic DNA analysis
analyzed case data from 2008 to 2010 that involved over 200,000 DNA anal-
yses (Kloosterman et al., 2012). The authors from the Netherlands Forensic
Institute noted that it was impossible to compare their results with other
studies despite the recommendation of the U.S. National Research Council
to conduct research to study the sources of error of various forensic disci-
plines. The council described two types of errors in forensic DNA testing:
Type 1 error: The DNA profile of a reference sample from a suspect is

concluded incorrectly to match with the crime sample.
Type 2 error: Wrongly reporting a non-DNA match between two sam-
ples when in truth a match exists.
The number of quality issue notifications constituted about 0.5% of the

casework samples examined. The authors noted an important difference
in impact between quality issues that have adverse outcomes on forensic
investigations and failures that are recognized and corrected in the early
stages of investigations.
The authors also stated that research into the rates of type 1 and
type 2 errors will provide only partial insight into the quality status of
a laboratory. The actual frequency of these errors is low and does not
allow recognition of trends in error rates in the forensic community. They

describe a system in place at the Netherlands Forensic Institute that relies
on failure and potential failure classifications.
8.7Overreliance on DNA technology

DNA profiling is a test, and like all scientific testing is subject to quality
control, even if automated systems such as robotics are used. Profiling is
also performed and interpreted by fallible human beings. Transcription
errors, handling mistakes, and other errors may contribute to incorrect
results. Case 1 from England shows that contamination may occur even
when automation and computers are used.
A common axiom is that computers do not make mistakes. However,
DNA analysis is still reliant on human beings at every step from isola-
tion of the crime stain to final interpretation (see Chapter3). Case 5 from
the United States describes transcription error and/or incorrect placing of
samples in vials and involves a DNA database match that subsequently
excluded the original accused.
Case 5
A masked man in a blue hooded sweatshirt burst into a wom-
ans home in Las Vegas in 2001 and forced her to drive to an
ATM for money. The man ran away when the womans hus-
band spotted them (Mower and McMurdo, 2011). Police fol-
lowed 18-year-old Dwayne Jackson and his cousin, Howard
Grissom, who were riding bikes, because the police thought
they could be the suspects.
The police looked inside a car in the driveway of the sus-
pects house and discovered a blue hooded sweatshirt with a
ski mask in the pocket. Jacksons DNA profile matched that on
the sweatshirt; the DNA was the only evidence connecting him
to the crime. Jackson pleaded guilty because the other charges
of kidnapping and burglary that carried lengthy terms would
be dropped if he did so. He was imprisoned for four years and
released in 2006.
In November 2010, the California Justice Department con-
tacted Las Vegas police and informed them that someone else
in the system matched the crime scene sample from the crime
for which Jackson had been convicted. Grissom was convicted
of an unrelated crime in southern California and was serving a
41-year jail term. His DNA profile matched the profile from the
Nevada crime. It was discovered through a forensic review and
r e-analysis that a laboratory technician put Johnsons sample

into Grissoms vial and vice versa in the original case. Jackson
was awarded $1.5 million in a settlement in July 2011.
The following bizarre case from England shows a consequence of tran-

scription error.
Case 6
Gareth Williams was a 31-year-old code breaker attached to
MI6 at the time of his death in August 2010. His decomposing
naked body was found inside a padlocked sports bag in the
bathtub of his flat in Pimlico, London over a week after he was
last seen (Davies, 2012). He had no injuries and no illicit or poi-
sonous substances were found in his body. The keys to the pad-
lock were beneath his naked body and the police determined
he could not have locked himself inside the bag.
A partial DNA profile belonging to a police scientist inves-
tigating the crime scene was discovered in February 2012. An
inquest in April 2012 revealed that a typographical error of a
forensic scientist in an email asking for a DNA check led police
to believe there was foreign DNA on Williams body. Detectives
wasted 18 months looking for a potential suspect who did
not exist. The coroner determined that a still unknown party
locked the victim inside his sports bag. The case is currently
undergoing a forensic review.
8.8Interpretation of DNA profiles:

Objectivity and subjectivity
Although DNA profiling is considered more objective than other foren-
sic disciplines, scientist discretion still comes into play, particularly in
interpreting partial DNA profiles or mixtures. Discretion is also a fac-
tor of crime scene selection, sample and item selection at the laboratory,
tests performed, and subsequent interpretation and communication of the
results. Crime scene samples are not pristine, and more often than not
provide less quantity and quality than a scientist would desire.
During the preparation of the appeal in the Jama case, the DNA sta-
tistics were recalculated by the forensic laboratory as a result of a review
(Vincent, 2010). The chance probability of a match changed to 1 in 150
million for the Australian Caucasian population (from 1 in 800 billion
originally) and to 1 in 17 million for the Somali population (1 in 89 billion

originally). Further calculations as a result of another review were likely
to yield results somewhere between these values. How could this occur if
the interpretation and statistical analysis of a DNA profile were objective?
The controversy concerning interpretations of DNA profiles by the
Victoria Police Forensic Services Department showed that interpreting
DNA profiles involves opinion and subjectivity (Fraser, 2010). Designating
a specific peak in a profile as true is subject to numerous factors, includ-
ing its appearance and ability to meet certain criteria. A true peak must
exceed a certain threshold value, have a certain height, be distinguishable
from artifacts, and allow explanations of peak imbalances. Including par-
ticular peaks in mixture calculations is also a subjective process.
One author describes subjectivity in interpretation of DNA profiles as
akin to the Texas Sharpshooter Fallacy or painting the target around
the arrow (Thompson, 2009). The criteria for determining matches or
inclusions for DNA profiles are purportedly shifted when a reference
sample of a suspect becomes known according to the fallacy. Post hoc
target shifting can distort the frequency and likelihood ratios, making
matches appear more probative than they actually are.
Epidemiologists named the Texas Sharpshooter Fallacy to describe
the tendency to assign unwarranted significance to random data by view-
ing it post hoc in an unduly narrow context (Gawande, 1999). As an exam-
ple, random cases of a disease may be interpreted to cluster in a particular
population due to some particular cause or effect.
The painting the target around the arrow description arose from
the story of a legendary Texan who shot his firearm randomly into the
side of a barn and then painted targets around the bullet holes. When the
paint dried, he invited his neighbors to see what a great shot he was. They
were impressed. They thought it extremely improbable that the shooter
could have hit every target dead center unless he was indeed an extraordi-
nary marksman and thus declared him to be the greatest sharpshooter in
the state. In summary, the evidence of his accuracy was far less probative
than it appeared.
This fallacy acts as a type of confirmation bias because the human
tendency is to interpret patterns in randomness where none exists. This
type of fallacy may be applied to partial or incomplete DNA profiles (low
levels and mixtures) subject to possible drop-out and d rop-in. Such pro-
files present potential for subjectivity. Thompson (2009) notes that a key
source of problems is the absence of formal standards for distinguishing
inclusions from exclusions.
An interesting study suggesting that DNA mixture interpretation
may be subjective was conducted on an adjudicated criminal case (Dror
and Hampikian, 2011). Seventeen North American expert DNA examiners
from one laboratory in the jurisdiction of the original case were asked
to interpret a mixture DNA profile and compare it to a suspect profile,
without any contextual information given to examiners in the original
trial. The experts came to conflicting conclusions about the inclusion or
exclusion of the suspect. Most of the context-free experts disagreed with
the original pretrial conclusions, suggesting to the authors of the study
that extraneous content influenced interpretation.
The DNA evidence related to a gang rape. One of the assailants testi-
fied against the other suspects in return for a lesser sentence. However,
those identified through the plea bargain denied involvement in the rape.
The original result was that the DNA of the suspects identified could not be
excluded from contributing to the mixture profile. The 17 examiners were
asked to review the DNA mixture profile, particularly that of Suspect 3,
originally determined as cannot be excluded in pretrial submissions.
The evidence presented to the 17 examiners consisted of electrophero-
grams relating to the sperm fraction from a vaginal swab and a reference
sample of Suspect 3. One examiner reported that the suspect cannot be
excluded, four examiners stated inconclusive, and 12 examiners said
exclude. This suggested to the authors of the study an element of sub-
jectivity in DNA interpretation. If the result was totally objective, all the
examiners should have reached the same conclusion, especially since they
all worked at one laboratory and used the same interpretation guidelines.
It is desirable that all subjective judgments about an electrophero-
gram be made without knowledge of the DNA profile of the person of
interest who contributes to a mixed DNA profile (Kelly et. al, 2012). This
should help prevent the raising of bias issues in court (Krane et al., 2008).
For example, an examiner should determine whether drop-out is possible
at a locus before looking at a reference profile.
8.9Retesting of samples
On occasion, it may be desirable to r etest samples, for example, if the pro-
cedures used by the initial testing laboratory are in doubt because of con-
tamination or some other error that occurred after an exhibit arrived at
the laboratory. Sufficient uncontaminated sample left on the item exam-
ined or sufficient uncontaminated DNA extract must be available for
retesting. During most testing protocols, sufficient DNA extract remains
for re-amplification by another laboratory. However, the sample remain-
ing on an exhibit may not be adequate for a retest, especially if touch DNA
is the suspect evidence.
Some jurisdictions require retention of part of a sample (DNA extract)
for further testing by the original laboratory. The volume remaining
should be indicated in the case notes. The DNA extract should be stored
according to the laboratory protocol. Laboratories should maintain freezer

systems specifically designed for DNA retention.
8.10Adversarial system
The Fraser team (2010) considered that different viewpoints are always
present in adversarial systems, and where possible the differences are
best resolved before trial. They noted that in Australia such pretrial meet-
ings were not held or were not fruitful. In the United Kingdom, a pretrial
review is a useful way to resolve disputes between defense and prosecu-
tion experts before any evidence is heard by a jury. The Fraser team felt
the most appropriate responses to court challenge consisted of whether
there was (1) active informed review of practices, (2) debriefing in prob-
lem cases, and (3) careful case preparation. The process of preparing for
trial and reviewing evidence must involve all partiesthe prosecution,
the defense, the police, and the scientists.
8.11Misconception about exact science

The idea that science can be exact and objective in the sense of detachment
from human judgment created many misconceptions about the nature of
DNA profiling (Evett, 1996). One misconception is that there is an exact
answer to the question of probability of a particular DNA profile given
that it came from someone other than the defendant. However, the prob-
ability of an event is inevitably conditioned by the assumptions made. It
is not possible in any situation to develop a probability without making at
least one assumption. As illustrated in the Farah Jama case (Chapter2) the
chance match involved a number of different probabilities.
Science is a complex subject involving many disciplines. Those
with interest in the philosophy of science can refer to Aitken (1995) and
Popper (1972).
8.12Obligations
Obligations are imposed on both the scientist who gives evidence at a crim-
inal trial and on the representatives of the legal system who are respon-
sible for conducting the trial. Justice Vincent (2010) reiterated this obligation
related to DNA evidence in his review of the Farah Jama conviction. This
obligation was enunciated also by a Royal Commission in Australia as early
as 1984 in a case involving fiber and paint evidence (Shannon, 1984).
A scientist should clearly and unambiguously describe the weight
and substance that should be applied to scientific tests and specifically
state the nature of any limitations.
The responsibility that rests upon the criminal justice professional is

to ask detailed and probing questions of the scientists in a case to elicit the
required information and convey it to a jury.
Chapter9 provides some ideas for questions that should be asked of
experts engaged by both prosecution and defense practitioners.
References
Aitken, C.C.G. 1995. Statistics and the Evaluation of Evidence for Forensic Scientists.
Chichester: John Wiley & Sons.
Beilby, V. 2006. Court issues concerning secondary transfer of DNA. Australian
and New Zealand Forensic Science Societys 18th International Annual
Symposium, Fremantle.
Daniel, R. and van Oorshot, R.A. 2011. An investigation of the presence of DNA on
unused laboratory gloves. Forensic Science International: Genetics, 3, e45e46.
Daniel, R. and van Oorshot, R.A. 2011. Environmental monitoring of background
DNA within a forensic biology laboratory. International Association of
Forensic Sciences 19th Conference, Madeira, September 1217.
Davies, C. 2012. Gareth Williams inquest hears of mystery DNA at crime scene.
Guardian, April 24.
Dror, I.E. and Hampikian, G. 2011. Subjectivity and bias in forensic DNA mixture
interpretation. Science and Justice, 51, 204208.
Evett, I.W. 1996. Expert evidence and forensic misconceptions of the nature of
exact science. Science and Justice, 36, 118122.
Fraser, J., Buckleton, J., and Gill, P. 2010. Review of DNA reporting practices by
Victoria Police Forensic Services Division. http://www.vicpolicenews.com.
au/images/stories/news/feature_story/victoria%20police%20forensic%20
services%20review%20%20report%20%20april%202010.pdf
Gawande, A. 1999. The cancer cluster myth. New Yorker, February 8, 3437.
Gill, P. and, Buckleton, J. 2010. A universal strategy to interpret DNA profiles
that does not require a definition of low-
copy-
number. Forensic Science
Gill, P., Rowlands, D., Tully, G.G. et al. 2010. Manufacturer contamination of
disposable plastic ware and other reagents: An agreed position statement
by ENFSI, SWGDAM, and BSAG. Forensic Science International: Genetics, 4,
269270, 2010.
Goray, M., van Oorshot, R.A., and Mitchell, J.R. 2012. DNA transfer within foren-
sic exhibit packaging: Potential for DNA loss and relocation. Forensic Science
Hadfield, S. 2011. Man sues state over DNA bungle. Herald Sun, November 12.
Hellmann, P. 2011. The H ellmann-Zanetti report on the acquittal of Amanda Knox
and Raffaele Sollecito. http://hellmannreport.wordpress.com
Himmelreich, C. 2009. Germanys phantom serial killer: A DNA blunder. Time
Magazine (available online).
Innocence Project. www.innocenceproject.org
Johnson, G. (State Coroner). 2006. Inquest into the death of Jaidyn Raymond Leskie.
Coroners Case 007/98, Melbourne, Victoria.
Kelly, H., Bright, J., Curran, J. et al. 2012. The interpretation of low level DNA mix-
tures. Forensic Science International: Genetics, 6, 191197.
Kloosterman, A., Quak, A., Szerps, M. et al. 2012. Errors in forensic DNA case-
work: What types, how many, how serious? Paper presented at conference of
European Association of Forensic Sciences.
Krane, D.E., Ford, S., Gilder, J.R. et al. 2008. Sequential unmasking: A means of
minimizing observer effects in forensic DNA interpretation. Journal of Forensic
Sciences, 53, 10061007.
Morris, S. 2012. Rape accused was victim of forensics error, regulator finds.
Guardian, October 10.
Mower, L. and McMurdo, J. 2011. Las Vegas police reveal DNA error put wrong
man in prison. Las Vegas Review Journal, July 8.
Popper, K. 1972. Conjectures and Refutations: The Growth of Scientific Knowledge.
London: Routledge.
Poy, A.L. and van Oorshot, R.A. 2006. Trace DNA presence, origin, and transfer
within a forensic biology laboratory and its potential effect on casework.
Journal of Forensic Identification, 56, 558576.
Rennison, A. 2012. Report into the circumstances of a complaint received from the
Greater Manchester Police, F SR-R-618, September 17.
Shannon, C.R. (Royal Commissioner). 1984. Royal Commission Report Concerning
the Conviction of Edward Charles Splatt. Adelaide Government Printer. http://
nla.gov.au/anbd.bib-an3292187
sharpshooter fallacy in forensic DNA interpretation. Law, Probability, and
Risk, 8, 257276.
Thompson, W.C., Taroni, F., and Aitken, C.G.G. 2003. How the probability of a
false positive affects the value of DNA evidence. Journal of Forensic Sciences,
48, 18.
van Oorshot, R.A., Ballantyne, K., and Mitchell, R. 2010. Forensic trace DNA: A
review. Investigative Genetics, 1, 14.
Vincent, Justice F. 2010. Inquiry into the Circumstances that Led to the Conviction of
Mr.Farah Abdulkadir Jama, Victorian Government Printer, Melbourne.
chapter nine
DNA pointers for criminal

justice professionals
9.1Introduction
Reviewing a complex forensic report to determine what questions should
be asked of the examiner and what areas must be challenged when dis-
cussing evidence in a criminal case is a daunting task. Ascertaining the
quality of the examination may demand considerable time and effort. A
full disclosure of laboratory records with a review by an independent
expert is a very common tactic in English and American criminal trials.
Sometimes experts are appointed by the court in European trials and act
for the court and are not adversary witnesses.
The complexity of DNA testing makes it difficult for a legal practitio-
ner to evaluate the evidence without expert assistance. Looking behind the
laboratory report to determine whether the underlying data support the con-
clusions should be the task of an expert witness. Experts may also assess
whether alternative theories (inadvertent transfer, laboratory error, or sam-
ple contamination) of the evidence have been presented or considered.
DNA profiles are complex and the propositions are also uncertain,
so the legal practitioner may ask how the statistical analysis should be
performed. An agreement regarding statistical analysis between the pros-
ecution and defense is beneficial and the statistical model should be able
to evaluate the differing positions.
The scientist is a facilitator of the discussion; strict boundaries must

be observed.
The scientist must clearly define and separate the issues of relevance.
He or she must eliminate confusion between the facts of the DNA
profile and the circumstances by which the DNA was deposited at
the crime scene.
The court then decides the relevant issues.
The next sections summarize the many issues discussed in this book and
may help a criminal justice professional understand and focus on DNA
evidence encountered in a criminal case.
147
9.2Advantages of DNA profiling

DNA profiling has a genetic basis and a firm scientific foundation.
DNA has the same composition throughout a body; DNA samples
from blood and other body fluids can be compared.
DNA is relatively stable and that explains its utility in cold cases.
DNA profiling can be used on minute samples, even samples such as
touch DNA that cannot be observed visually.
DNA profiling has very high discrimination potential and thus the
power to exclude an individual is its fundamental advantage.
DNA profiling can provide high statistical value when a match is
found; this lessens the possibility that another person has the same
DNA profile.
9.3Querying DNA evidence: Advice

for the prosecution and the defense
When do you call a forensic DNA prosecution expert to testify? If the
prosecution calls an expert it is prudent to discuss the case with the expert
to determine any strengths or weaknesses in the case. The prosecutions
failure to take this step may allow the defense to expose weaknesses of
the results. Both parties should ensure that their experts are open and
frank. Experts should be asked about possible quality failures, contamina-
tion, avenues of transfer, and inadvertent transfers.
Audit trails and quality assurance procedures that prevent contami-
nation may be readily apparent to an expert in the field. The details can
also be explained privately to the lawyer in the case rather than in a court-
room. The lawyer may also be more prepared to examine the opposi-
tions forensic witness after discussions with his or her own expert. Issues
discussed in this text, for example, hypothesis testing, subjectivity, and
limitations, should always be considered by a presenting or challenging
lawyer if forensic science evidence is crucial to a case.
Cases involving mixtures of DNA require discussion between the
prosecution and the defense to determine how many contributors may be
represented in the mixture. The defense counsel and their experts should
also discuss this issue. Usually the smallest number of contributors is pro-
posed by the prosecution in order to explain the evidence and maximize
likelihoods. If the number of contributors is vital to the case, it is wise to
prepare and review the calculations to prevent delaying the court pro-
ceedings to make the calculations.
Cellular material of an unspecified body origin, such as touch DNA,
will need more consideration than DNA readily associated to blood,
semen, or saliva. The modes of deposition must be postulated. The time
Chapter nine: DNA pointers for criminal justice professionals 149
of deposition of DNA cannot be determined and that fact may or may not
be critical to a case.
If a haplotype is cited in a report, it refers to Y
-STR or mitochondrial
DNA profiling, not nuclear DNA profiling. Review the body of the report
or any appendices for an explanation of the technique. The statistics in
these types of profiles are much lower because of the inheritance of the
DNA through the paternal or maternal line. Y-STR and mitochondrial
techniques are still very useful for exclusionary purposes.
Technical problems may pose a major issue in a case. An expert may
determine whether the controls used in the analysis accord as expected or
present problems. What does a DNA profile (electropherogram) look like?
Is it of good quality with all peaks above the stochastic threshold? Or is its
quality poor, displaying many artifacts, overloading of DNA, inhibition,
or drop-out?
Are the limitations of the techniques used described in the report and
testimony? If not, why not? This book explains in detail the importance
for forensic scientists to communicate both the limitations and advantages
of DNA profiling.
The collection methods utilized at crime scenes should be rigorous.
Several cases in this book illustrate the pitfalls of poor collection practices.
The appropriate personal protection equipmentgloves, face masks, hair
nets, disposable overalls, and overshoesshould be used at all times. The
gloves should be changed during collections of exhibits to prevent the
ready transfer of DNA discussed previously in this book.
The chain of custody must be complete and conform to anti-
contamination measures in place. Evidentiary items for submission to
DNA analysis should be bagged individually to prevent the transfer
of DNA between exhibits within a bag and transfer of DNA from one
part of an exhibit to the other (see Chapter8). Revisiting a scene to col-
lect exhibits may be unavoidable, but the investigator should assess scene
security between visits.
9.4Warning signs
A number of warning signs may alert a legal practitioner when examin-
ing a case involving DNA evidence.
Is one DNA result the only evidence in the case? Guidelines are in
place around the world to ensure that no prosecution should occur if the
sole evidence in a case comes from DNA. The Farah Jama case in Australia
(Case 4, Chapter 2) and the Adam Scott case in England (Case 1, Chapter 8)
show the consequences of unquestioned reliance on DNA. These individ-
uals were imprisoned when in fact no offenses occurred. The DNA results
of every case must be considered very carefully. A legal practitioner
should consider the DNA result, how the DNA was deposited, and how
the DNA result fits in with the other circumstantial evidence in the case.
The more DNA results of evidential value, the less likely contamina-
tion may have taken place. This is especially true if different body fluids
or t wo-way transfers are involved. Consider a case in which a DNA pro-
file from blood found on the suspects clothing matches the victims DNA
profile and DNA from semen matching the suspects DNA is found on the
victim. Contamination would be considered a much smaller possibility in
this case than it would be in a rape case involving a single medical swab
from a victim containing semen matching the accused. Finally, could the
DNA have been deposited through legitimate contact, using the princi-
ples of trace evidence transfer?
Is the DNA relevant? If the accused and the victim cohabit, DNA
transfers would not be surprising. What is more pertinent is the ability to
associate DNA with a body fluid such as blood or semen.
Cold cases are often re-examined, but the exhibits may have been ini-
tially examined in conditions lacking the strict contamination mitigation
measures used today. The reason is that transfers of minute quantities of
DNA were not considered before the 1990s.
The probative values of partial profiles, mixtures, and low level sam-
ples are particularly difficult to interpret and understand. Some reports
do not state that a sample contained low level DNA. Low statistical val-
ues may or may not be an indication of low level DNA. Ask your expert
whether a DNA profile is suboptimal. Could the DNA have been degraded
or inhibited so that only a partial profile was obtained? Who are the con-
tributors other than a suspect in a mixture of DNA? Were all necessary
reference samples obtained so that they could be eliminated from the pro-
file? Is masking an issue? Could a major contributor be separated out or
were statistical packages used to deconvolute the contributors?
9.5Was all evidence tested?

There may be, for some reason, evidence that may be relevant or pertinent
to the case that was not tested. Evidence may not have been collected from
a crime scene or victim, or it may have been collected but not tested. The
Frank Button case from Australia (Case 8, Chapter 2) provides an example;
again the defendant served time in prison until another individual was
proved to have committed the crime.
A decision may be made not to test an item because of financial fac-
tors, time, sufficient evidence already obtained, limited resources, or
direction from the submitting agency. Also a decision may be made to test
only one stain. One semen stain on a sheet may be tested even if many
were observed on the item.
9.6Pretrial review
A pretrial review should include all partiesprosecution, scientists,
police, and defense team. Differing viewpoints are best resolved before
trial. It is also beneficial if the parties agree to certain matters relating to
DNA interpretation (e.g., numbers of contributors to a mixed DNA profile)
before trial. The defense may request that its own expert re-analyze the
samples. If insufficient DNA extract remains from the sample in the origi-
nal laboratory, it may be possible to examine the original exhibit and
attempt to obtain additional DNA. However, all extract is often consumed
in the original testing, especially in testing touch DNA.
Debriefings of problem cases should be conducted by the prosecution,
the defense, and the forensic laboratories.
9.7Suggested cross-examination questions

Before the trial (soon enough to allow an expert sufficient time to review
the data), the legal professional should request full disclosure of labora-
tory records including case notes and methods used. Suggested questions
are categorized and listed below.
9.7.1General
Were the collection policies and practices at the crime scene or medi-
cal examination optimal in the analysis of this case?
Was a rationale for testing explained in the notes and or statement?
If not, what was the rationale?
Was the scientific method used (and what is that?)
Have alternative hypotheses been considered? What are they?
Why is DNA profiling so powerful? (It has a high discrimination
power and the power to exclude.)
Have the meanings of the scientific terms used been properly explained?
Was an i mpact-based priority testing system used?
What quality assurance procedures were in place?
Is the examiner aware of observer and/or context effects?
Does the examiner know the error rates of the tests? Can he or she
explain this concept?
How have the statistics quoted in the report been determined?
Is there a possibility of transfer (primary, secondary, or higher)?
Did the positive and negative controls perform as expected?
Does the laboratory have databases for investigating contamination
events including elimination databases for consumable suppliers
(where possible), police officers attending crime scenes, crime scene
operators, and laboratory staff (scientific and administrative)? What

are they specifically?
Were there issues with the technical review?
Were there issues with the administrative review?
9.7.2Single source DNA profiles associated

with blood, semen, or saliva
Can the DNA profile be related to a specific body fluid? If so, how?
What reference profiles were used and how were they obtained?
Were all appropriate reference samples taken and profiled?
Was the evidence profile interpreted and designated as single source
before comparison with reference DNA profiles?
How can we be sure that contamination was prevented?
What are the limitations of the results?
Have the appropriate population databases been used?
Was extra scrutiny applied if there is only one DNA result from many
items tested?
If there is an inclusion, or a match, what is the statistic and what does
it mean?
9.7.3Difficult DNA profiles (partial, low level, mixture,

unspecified origin)
Is this a partial DNA profile and why?
Does this DNA profile exhibit degradation or inhibition and why?
Was the sample re-amplified to obtain a better result? If not, why not?
Are any of the samples mixtures from two or more individuals?
Can the mixtures be separated into major and minor contributors,
and if so why and how?
What are the possible methods of transfer of the DNA?
Can the DNA detected be related to a particular time?
Can the DNA be related to a particular body matter? If so, how?
Do any profiles exhibit low level DNA and require extra scrutiny?
Are any of the peaks in the profile below the stochastic threshold?
How has the witness dealt with this? Would the witness say this
profile is suboptimal?
If the profile is low level, what extra precautions were taken, if any?
How was the final profile derivedthrough consensus profiles or
the statistical model? What was the rationale?
9.7.4Expert witness
Does the witness have an appreciation of DNA interpretation prac-
tices internationally? (References cited in this book may be useful
for preparing questions.)
Does the witness participate in a continuing education program?
9.8Discovery requests
This section summarizes the documents required for disclosure of scien-
tific materials pertaining to DNA testing. The summary is applicable to
all DNA testing that has been performed, is in progress, or planned for
the future. Figure3.6 in Chapter3 lists the documentation that should be
included in every case file.
Case fileDocuments listed in Figure3.6 including all records gener-
ated by the testing laboratory and copies of all photographs should be com-
piled and available for discovery. Case-related correspondence between
investigating police, other officials, and staff members should be included.
Laboratory protocolsCopies of all standard operating protocols
used in connection with laboratory testing should include explanations
of the method of DNA analysis, the statistical interpretation details, perti-
nent validation studies, population databases, and allele frequencies.
Chain of custodyAll records that document the handling of the
evidence from the initial point of collection to current disposition should
be retained for discovery. The records should indicate how the materials
were stored (temperatures and types of containers), the amount of evi-
dence consumed in testing, the amount of material remaining, and where
and how the remaining evidence is stored.
SoftwareA list of all commercial software programs used in the
DNA testing should include name, manufacturer, and version used.
Data filesData describing extraction, quantification, amplification,
separation, and analysis should be maintained in case files. The data
should indicate the dates when steps were performed and names of per-
sonnel performing tests and checking results.
STR frequency tablesCopies of allelic frequency tables used in
making statistical estimates should be retained. If the testing laboratory
relied upon published data, this requirement may be satisfied by specific
references to sources.
Unintended transfer or sample contaminationRecords maintained
by the testing laboratory should document instances of unintended trans-
fer or sample contamination and describe all corrective measures taken.
AccreditationIs the laboratory accredited? Has this accreditation
ever been suspended? If so, for what reason?
Appendix A: Glossary of terms
used in reports and testimony
Allele: Alternative at a site on a DNA molecule; one alternative is inher-

ited from each parent; variation at a given locus on a chromosome;
number of short tandem repeats at a locus on a chromosome.
Allele drop-in: Contamination from a source not associated with a crime
stain and manifested as one or two alleles.
Allele drop-out: Low level of DNA insufficiently amplified to yield a
detectable signal.
Amelogenin: Part of DNA strand used to characterize the sex of an indi-
vidual contributing DNA.
Amplification: Process by which the number of copies of specific DNA
sequences is increased via sequential copying.
Analytical threshold: Minimum height requirement at and above which
detected peaks can be distinguished reliably from background
noise on an electropherogram.
Artifact: Result appearing in DNA profile (electropherogram) that arises
from the process and is not intrinsic to the DNA tested; also see
stutter and pull-up.
Autosome: Chromosome not involved in sex determination; humans
have 22 pairs.
155
156 Appendix A: Glossary of terms used in reports and testimony
Base pair: DNA is formed from four chemical bases; a base pair (bp) con-
sists of a base in one strand of the double helix and its comple-
mentary base on the other strand.
Buccal swab: Mouth swab used to obtain DNA samples.
Chromosome: Discrete unit of the genetic material carrying genes; chro-
mosomes are arranged into structures that can be visualized dur-
ing cell division.
Co-ancestry coefficient: Allowance for possible shared ancestry within
a population; designated between 0 and 1; higher values corre-
spond to greater shared ancestry.
Confirmatory test: Test used to confirm the presence of a specific biologi-
cal material such as blood or semen.
CODIS: Combined DNA Index System database for DNA profiles used
in the United States.
Conservative: (1) Assignment for weight of evidence that is believed to
favor the defense. (2) When the evidence is very powerful in one
direction, assigning a weight below the level of belief in that direc-
tion. (3) Lack of conservativeness often results when the assump-
tions underpinning a statistical model are seriously violated.
Contamination: Extraneous DNA from a source not associated with a
crime stain, for example plasticware can be contaminated at a man-
ufacturing source.
Crime scene sample: Sample taken from a crime scene or body by crime
scene investigators or medical personnel.
Degradation: Breakdown of DNA strands through age, environment, or
chemical insult resulting in a greater loss of longer fragments.
Deoxyribonucleic acid (DNA): Chemical compound found in all nucle-
ated cells of a body. It codes for characteristics in humans.
Electropherogram: Output of electrophoresis instrument that displays
DNA profiles as peaks on a graph.
Electrophoresis: Method of separating molecules based on size and
charge, used to separate DNA fragments of varying length by
application of an electric current.
Exclusion: Exclusion of a contributor to stain. (1) Decision by an expert that
a certain reference DNA profile does not represent a contributor to
the stain. (2) Situation in which a reference profile is excluded.
(3) Exclusion from a stain at one or more loci. (4) Exclusion at a
locus. (4) Pattern of assumed genotypes at a locus indicating that
an allele seen in a certain reference DNA profile is not observed
in a stain.
Exclusion probability: Probability of the exclusion of a randomly selected
DNA profile.
Appendix A: Glossary of terms used in reports and testimony 157
Familial searching: Process allowing potential relatives of offenders to

be identified in a DNA database when the offender profile is not
in the database.
Frequency: Rate at which an event occurs. For example, sample frequency
of an allele is the number of occurrences of the allele in a popula-
tion sample divided by the sample size: population frequency of
a DNA profile is the (unknown) number of times that the profile
occurs in the population divided by the population size.
Gene: Site on a DNA molecule; sequence of the inherited code for which
there is a functional product; sequence of DNA base pairs.
Genome: Total genetic material of an organism contained in a full set of
chromosomes.
Genotype: Characterization of alleles at a specific site; the designation of
two alleles at a locus is a genotype.
Haplotype: Collective genotype of a number of closely linked loci on a
chromosome that are inherited together; in mitochondrial DNA,
the sequence of the control regions that pass unchanged from a
mother to her offspring; in Y-STR typing, a Y-STR sequence that is
inherited unchanged from male to male.
Heterozygote: Two alleles of an individual at a particular site are different.
Homozygote: Two alleles of an individual at a particular site are the same.
Intelligence-led screen: Screening conducted during major crime inves-
tigations applied to a DNA sample from a number of people who
may have been associated with a crime or crime scene with the
intent to eliminate them from the inquiry.
Likelihood: Conditional probability of an event; an event is consid-
ered an outcome corresponding to one of several conditions or
hypotheses.
Likelihood ratio (LR): Ratio of two probabilities of the same event under
different hypotheses. The numerator typically contains the pros-
ecutors hypothesis and the denominator is the defense hypoth-
esis. The LR is often expressed as the ratio between the likelihood
that a given profile came from a particular individual and the
likelihood that it came from a random unrelated person.
Locus: Area of DNA that is analyzed to generate a DNA profile. Loci is
the plural.
Low level or low template analysis:Analysis result from very small
amounts of DNA that typically produces peaks below the sto-
chastic level.
Masking: Result from analysis of a DNA mixture showing overlap of the
same allele originating from different contributors.
Mitochondrial DNA: DNA obtained from the mitochondria of a cell.
158 Appendix A: Glossary of terms used in reports and testimony
Mitochondrion: Subcellular unit within the cell that provides energy.

Mitochondria is the plural.
Mixture: DNA typing result originating from two or more individuals.
Mixture ratio: Relative ratio of DNA contributions of multiple individuals
to a mixed typing result based on quantitative peak height infor-
mation. A mixture ratio may also be expressed as a percentage.
Molecule: Chemical substance consisting of atoms bound together in a
specific structure.
Multiplexing: Method of amplifying multiple sites on a DNA molecule
in a single reaction vessel.
Nuclear DNA (nDNA): DNA found in cell nucleus. Nuclear DNA testing
consists of autosomal STR DNA typing and Y-STR DNA typing.
Nucleus: Cell structure that contains most of the DNA. Nuclei is the plural.
Partial profile:Result obtained from a sample deficient in quality or
quantity; no full profile is produced.
Polymerase chain reaction (PCR): DNA amplification process by which
one or more DNA regions are copied using a DNA polymerase
enzyme to generate enough DNA for analysis.
Presumptive test: Screening procedure that indicates the possible pres-
ence of a specific body fluid.
Primer: Reagent containing synthesized DNA sequences that are com-
plementary to a specific segment of the DNA on either side of the
area of interest used in the PCR process. Each primer is labeled
fluorescently so that the copied STR allele can be detected visu-
ally and its length measured
Probability: Rate of occurrence of an event in a repeatable experiment;
expected frequency; number between 0 and 1 that reasonably
reflects the belief that an event is true.
Profile: Numerical format showing STR alleles detected; one or more
genotypes used for DNA comparison.
ProfilerPlus: Technique used in Australia to profile DNA samples. It ana-
lyzes 10 areas of a DNA molecule including the sex chromosome.
Proposition: Hypothesis of the defense or prosecution arguments used
to formulate a likelihood ratio.
Pull-up: Artifact resulting from fluorescent detection.
Random match probability (RMP): Probability of randomly selecting an
unrelated individual from a population who has the same DNA
profile as that of the questioned sample.
Relative fluorescent unit (RFU): Unit of measurement of peak height on an
electropherogram resulting from detection of fluorescence intensity.
Restricted combinatorial method: Elaboration of the unrestricted method
in which allelic intensity (peak height/area) data are used to restrict
the sets of genotypes that are considered plausible explanations.
Appendix A: Glossary of terms used in reports and testimony 159
Resolvable DNA mixture: Mixture of DNA from two or more individu-

als detected on an item of evidence in which the ratios of major
and minor contributors can be deduced due to the proportion of
one versus the other.
Sex chromosome: Chromosome involved in determining the sex of an
individual. Females possess two X chromosomes and males pos-
sess one X and one Y chromosome.
Short tandem repeat (STR): Type of locus used to generate DNA profiles
through nuclear typing.
Stochastic effect: Imbalance in the amplification of two alleles by com-
petition during PCR; one allele is preferentially amplified over
the other.
Stochastic effect, random: Variations in alleles on repeat sampling with
small amounts of DNA.
Stochastic threshold:Minimum DNA quantity required to produce a
reliable or optimum profile, assuming that d rop-out of a sister
allele has not occurred at a particular locus above this threshold
determined in specific laboratory studies.
Stutter: Allelic artifact caused by slippage of the polymerase enzyme
which is four bases fewer than the allele that causes the stutter.
Over-stutter (one base pair longer than its associated allele) occa-
sionally occurs. Stutters are always found in allelic positions and
can compromise interpretation of minor contributions to mixtures.
Touch DNA: DNA from skin (epithelial cells) that is left behind when a
person touches or otherwise comes into contact with an item or
other person.
Unrestricted combinatorial method: Simple likelihood ratio method of
evaluating mixture evidence. The method assumes a list of all
alleles in the mixture and considers competing hypotheses that
various known or unknown profiles are constituents of the mix-
ture. It uses no data on allelic intensities, hence one set of genotypes
whose allele sets are coincident with the mixture is considered to
have as valid an explanation of the mixture as any other set.
Y-STR profiling: DNA typing in which STRs on the Y (male) chromo-
some are analyzed.
References
Gill, P., Brenner, C.H., Buckleton, J.S. et. al, 2006. DNA Commission of the
International Society of Forensic Genetics: Recommendations on the inter-
pretation of mixtures. Forensic Science International, 160, 90101.
Puch-Solis, R., Roberts, P., Pope, S. et al. 2012. Practitioner Guide 2: Assessing the
Probative Value of DNA Evidence. London: Royal Statistical Society, London.
Appendix B: Selected DNA
issues and case examples
Chapter Case Description

Notable cases solved by DNA
1 1 Colin Pitchfork case; first murder solved by DNA profiling;
first DNA-led screen; first person exonerated by DNA
4 Kirk Bloodsworth case; first conviction overturned by DNA
testing
2 1 Ray Krone case; contrast of DNA with questionable science
2 Damon Thibodeaux case; questionable assumptions about
absence of evidence
3 Robert Dewey case; stability of DNA
3 3 Darren Paul Smith case; cold case solved through DNA
database
5 Analysis of DNA from maggots ingesting body
6 2 Mass screening through Y-STR analysis
7 1 Paul Ware case; mitochondrial DNA
2 Sedrick Courtney case; mitochondrial DNA exclusion of
convicted individual
5 Lynette White case; familial DNA searching
6 Grim Sleeper; familial DNA searching
7 Shirley Duguay case; analysis of domestic animal hair
continued
161
162 Appendix B: Selected DNA issues and case examples
Chapter Case Description

Statistics
2 5 Twins
4 1 Ayturgrul v. The Queen; mitochondrial DNA
2 Patrick Waring case; inclusion and exclusion
3 Timothy Durham case; mixture erroneously treated as single
source
5 3 Josiah Sutton case; mixture interpretation
4 Case from authors files; complex mixtures
7 3 Scott Peterson case; mitochondrial DNA statistics
Partial profiles and low levels

2 7 Murder of Meredith Kercher
5 1 Omagh bombing; low level DNA
Time of deposition
2 6 Case from authors files
Relevant evidence
8 Frank Button case
3 1 Damilola Taylor case
2 Case from authors files
Poor collection practices and compromised evidence

1 2 O.J. Simpson case
5 1 Omagh bombing
Contamination
2 4 Farah Jama case; contamination during medical examination
8 1 Adam Scott case; contamination during extraction step of
analysis
2 Jaidyn Leskie case; contamination between samples from
different cases
3 Murders of Margaret and Seana Tapp; contamination
between samples from different cases
4 Phantom of Heilbronn; contamination of sample
consumables
Errors
8 5 Dwayne Jackson case; switching of DNA samples in vials
6 Gareth Williams case; typographical error
Appendix C: Steps in review
of evidence
C.1Obtaining further information

from documents and client
One reason for requiring further information may be the possibility of
innocent contact or other transfer that may explain a DNA match.
C.2Review of scientific reports

In the course of a review, certain questions should be asked in an effort to
clarify the report contents:
What is the nature of a sample? Is it blood, semen, cellular material?

Has the nature of the material or the body origin been confirmed
through confirmatory testing?
Is the sample relevant to the matter in contention?
What is the quality level of the sample?
How many pertinent DNA results are there?
Is there a possibility or probability of contamination?
Define the strengths and limitations of the evidence even if they are not
detailed in reports.
163
164 Appendix C: Steps in review of evidence
C.3Requesting materials for review

The case file notes (see Figure3.6 in Chapter 3) should include the follow-
ing items but a specific case may require additional documentation.
Formulation of hypotheses
Testing plan and sampling rationale
Photographs or sketches taken in the laboratory
Examination or bench notes made by laboratory analyst
Electropherograms of DNA profiles including reference profiles
Designations of alleles in electropherograms
Thresholds used in designation of alleles
Statistical calculations and description of population used
Conclusions reached and bases for conclusions
Standard forms related to testing such as batch numbers of reagents
used, method, and equipment
Communications between the analysts and others involved in the case
Checks of results against staff DNA database
Quality assurance (technical and administrative) reviews
All case reports issued by laboratory including preliminary reports or
notes of verbal examinations
Certain documents that may not be included in a case file but may be
needed for review and possible future use:
Laboratory protocols such as DNA interpretation standards and statis-

tical methods
Records relating to the chain of custody
Details about storage of materials
Descriptions of software programs
Equipment maintenance and calibration records
Analyst training and proficiency test records
Unexpected results or corrective action reports
Quality assurance reports and audits
C.4Examination by independent expert

Always consult with your scientific expert before requesting material
from the opposing party. The expert should examine all reports and raw
data. He or she should also clear up any potential bias on the part of the
opposing partys expert.
Index
A Fraser review, 86, 88
legal cases in, 9, 2021, 26, 29, 41, 7374,
ABO blood grouping, 5, 16, 33 123, 124125
Acquittals, 18, 27, 29, 39, 134 Mungo man study, 125
Admissibility, 7, 8, 9, 69, 123 Autosomal STR profiling, 4, 17, 4546, 107,
Adversarial systems, 144 114; see also DNA analyses
Aitkin, C., 65 differential extraction, 94, 107
Alleles, 1, 17; see also Electropherograms inconclusive results with, 51, 74, 76,
appearance, 66 105106
drop-in/out, 83, 85, 88, 99 main steps in, 45
frequencies, 68, 153 statistical significance with, 70
nonconcordant, 99 Autosomes, 114, 129
off-ladder, 57 Avery, Oswald, 2
reading tables of, 47, 48, 49, 50 Ayturgrul v. The Queen, 2012, 69, 123
Amelogenin, 17, 47, 155
AmpFLSTR Identifiler, 22, 129
AmpFLSTR Profiler Plus, 22
B
AmpFLSTR SGM Plus, 22 Bacteria, 129
Amplification, 53, 84, 90, 129 Balding, D.J., 69, 76
controls, 51 Base pairs, 3, 46
procedure, 53 Bases, 5
sensitivity threshold, 84 Bayes theorem, 6869, 112
with small amounts of DNA, 81, 84, 99 Beamish case, 128
and stochastic variation, 86 Belfast bombing, 84
Analytical threshold, 6061, 96, 99 Bias, 74, 76, 98, 142, 143
Ancient DNA studies, 14, 24, 117, 125 Binary matches, 71
Andrews case, 7 Biological model, 91
Animal hair, 128129 Biologist Special Advisory Group (BSAG)
Archaeological findings, 118, 125 -Australia, 137
Armed robbery case, 2627 Birthday problem, 24, 71
Artifacts, 55, 57, 58, 59 Bite marks, 1415
dealing with, 88 Blind testing, 139
pull-up, 57 Blood evidence, 34, 3738, 41
stutter, 55, 56 confirming, 36
Audit trails, 133, 135, 148 cross-examination questions for, 152
Australia degradation, 38, 118119
aboriginal population, 26, 77 locating, 36
database, 22 reference samples, 44
165
166 Index
Blood grouping, 5, 16, 33 with low level DNA, 9293

Blood relatives, 68, 7778, 109 of mitochondrial DNA samples,
Bloodsworth case, 10 124125
Body fluid testing, 36 prevention, 135
Bone testing, 7, 117119, 124, 125 in rape cases, 131, 134, 135
Brother defense, 77 sources of, 135, 136
Buccal swabs, 44, 52 Courtney case, 19, 121
Buckleton, John, 70, 86, 88, 144 Crick, Francis, 5
Burnt remains, 44, 119 Crime scenes; see also Collection;
Butler case, 109 Contamination
Button case, 29, 133134, 150 mistakenly linking, 136137
outdoors, 81
C security, 149
transfer by personnel, 133, 136
Capillary electrophoresis, 51, 54 CrimTrac National Criminal Investigation
Case files, 47, 61, 62, 153 Database (NCIDD), 22
Castro case, 7 Cross-examination
Cats, 128129 for difficult profiles, 152
Chain of custody, 149, 153 of expert witnesses, 153
Chemoluminescence, 36 general questions, 151152
Chromosomes, 2, 5, 17 for single source DNA of specific body
Clayton, T., 70, 98 fluid, 152
Clothing samples, 34, 41, 134135 CSI effect, 30
Co-ancestry coefficient, 77
Cold cases, 134, 137 D
re-examining exhibits, 19, 126, 131, 150
solved through DNA, 18, 40, 4142 Dandruff, 40
Collection Data files, 153
contamination during, 93, 133 Data sharing, 18
instruments and transfer, 136, 137 Database searches, 23, 132, 133
personal protection equipment, 149 Databases, 15, 71, 110; see also Populations
procedures, 8, 29, 84, 93, 149, 150 comparison types, 23
Combined DNA Index System (CODIS), 15, core STR loci, 22
22, 46, 101 information stored in, 22
Combined probability of exclusion (CPE), international, 123
6970; see also Exclusion national, 22
Combined probability of inclusion (CPI), 69 size, 68, 76
Computer software, 60, 70, 95, 153 validation, 70
Confessions, 6, 15, 107108 Death row, 10, 15, 122
Consensus strategies, 86, 91 Debriefings, 144, 151
Contamination; see also Transfer Declared contributors, 7576
burden of proving, 135 Deen case, 72
between cases, 132, 134 Defense
cleaning techniques, 59, 124, 125 experts for, 148149, 151
of clothing exhibits, 134135 fallacy of, 7173
in cold cases, 131 hypothesis, 67
documenting, 153 questioning DNA evidence, 30, 148149
due to collection practices, 93, 133 responsibilities, 145
identifying source of, 135 Degradation, 19, 24
in laboratory, 133, 134 of blood evidence, 38, 118119
legal cases involving, 2021, 93, 132133, exposure to elements, 18, 81, 119
134, 135, 136137 factors affecting rate of, 18
Index 167
fire damage, 119 decontamination, 124125

observed on electropherograms, 57, 58, endogenous, 125
82 mishandling, 8, 2728, 84, 133, 140
resulting in partial profiles, 81 persistence of, 2527
of saliva, 40 preservation, 18
ski slope effect, 8182 quantity, 10, 25, 3435, 37, 82, 84
Deoxyribonucleic acid (DNA) molecule, 33 querying at trial, 148149
in cellular makeup, 16 relevance, 9, 25, 2728, 29, 37, 3839
and concept of individuality, 45 retention/storage, 18, 143144
and law of inheritance, 1 searching for, 3537
structure, 23 as sole evidence in case, 131, 132, 133,
variable numbers of tandem repeats, 5 140, 149
Deposition; see also Transfer stability of, 1819
modes of, 26, 137, 148, 150 testing all collected evidence, 29, 37,
through legitimate contact, 132, 150 133134, 150
time of, 2527, 43, 149 of unspecified body origin, 34, 93, 138,
Dewey case, 18, 19 148, 152
Dinosaur DNA, 125 value of, 9
Disclosure, 147, 153 DNA extract, 18, 143, 151
Discovery requests, 153 DNA fingerprinting, 5, 7
DNA, see Deoxyribonucleic acid (DNA) DNA mixtures
molecule analysis steps, 98
DNA analyses; see also Amplification; complex, 95, 100101
Interpretation contributors to Y-STR profile, 113
automation of, 140 cross-examination questions, 152
batch processing, 51, 60, 132 discussion between opposing counsel,
of bone, 117119 148
case documentation, 61, 62 four or more contributors, 95, 100101
confirmatory tests, 34, 36, 138 interpretation, 96, 98, 113, 138, 142143
controls, 5152, 149 legal cases involving, 75, 94, 9596, 100
detection, 54 likelihood ratios, 97, 99, 100, 113114
extraction, 52, 94 low template, 99
identifying body fluid, 34, 36 major and minor contributors, 28, 95, 97,
objectives of, 3536 113, 124
peak designation, 6061 male and female, 106, 107
processing steps, 51, 132 of mitochondrial DNA, 124, 125126
quantification, 5253 mixture ratios, 113114
quantity of extract, 151 two or more people, 9397
retesting samples, 143144 unknown number of contributors, 98
sensitivity, 90 unrestricted versus restricted
separation, 54 calculations, 97
from single cell, 10, 25 victim DNA, 36
technical issues, 55, 56, 57, 58, 59 DNA profiling, 5, 7, 9; see also DNA
template thresholds, 84, 86, 90 analyses; Mitochondrial DNA
time required for, 5960 (mtDNA); Y-STR profiling
DNA evidence; see also Collection; advantages of, 14, 76, 148
Contamination; Degradation; biological materials allowing, 3335
Deposition; Transfer discrimination power, 6, 14, 1516, 65
admissibility, 7, 8, 9, 69, 123 fallibility, 13, 30
bagging, 138, 149 genetic basis for, 1618
confirmatory tests, 34, 36 kits, 17, 22, 23, 84, 129
in context, 2425, 131 misconceptions, 3, 144
168 Index
public perception of, 3, 15 Exhibits; see also DNA evidence

techniques used, 4, 17, 105, 128, 129, 149 relevance, 29
DNA replication, 5 searching for DNA on, 3537
DNA sequencing, 34 transfer between, 138, 149
DNA technology Exonerations, 10, 13, 15, 73, 96, 139
early court challenges to, 79 Experts, 30, 147
improvements in, 910 alternative theories, 73, 98, 147
overreliance on, 140141 conflicting conclusions by, 143
Documentation, see Case files; Reports cross examination questions, 153
Dogs, 125, 128, 129 explaining limitations, 73, 131, 144
Dominant genes, 1 obligations of, 131, 144
Dragnets, 23 subjectivity, 141143
Drop-in/out, 83, 85, 86, 88, 99 when to call to testify, 148
Duguay case, 128 Eye color, 44
Duke of Edinburgh, 117
Durham case, 75 F
E Fallacy of the transposed conditional,

7173
Electropherograms False positives, 75, 138, 139
artifacts, 55, 56, 57, 58, 59 Familial matching
experiment involving, 8890 basis for, 127
indications of degradation, 57, 58, 82 legal cases, 126127, 127128
measurement units, 54 with mitochondrial DNA, 126128
partial profiles, 58, 83 permission policies, 127
peak designation, 6061, 66, 89, 99 of Romanov family, 117
reading, 5455 search technique, 24, 126, 127
samples, 49, 50, 83, 87, 89 Feces, 43
ski slope effects, 59, 8182 Fiber evidence, 144
stochastic threshold, 47, 61, 90 Fingernail scrapings, 40
Electrophoresis, 51, 54 Firearms analysis, 127, 136
Emerging techniques, 4344 Forensics, 1314, 30
EMPOP 9 database, 123 Franklin, Rosalind, 2, 3
Enhanced interrogation approach, 86 Franklin case, 127128
Enhancement techniques, 84, 90 Fraser, Jim, 86, 88, 144
Enzyme typing, 5 Fraser review, 86, 88
Epithelial cells, 34, 36, 41 Frequencies, see Statistics
Error rates, 138140 Full disclosure, 147, 151
Error types, 139
Ethnicity, 68, 77 G
European Network of Forensic Science
Institutes (ENFSI), 137 GeneMapper, 60
Evett, Ian, 76 Genes, 1, 2, 3, 4, 17
Evidence, see DNA evidence Genescan, 54
Exclusion, 9, 14, 1516, 22, 149 Genomes
combined probability of, 6970 full human, 34
expert disagreement on, 143 mitochondrial versus nuclear, 119
percentage, 123 Genotype frequencies, 67, 68, 70
standards for, 142 Genotyper, 54, 60
as used in forensic statements, 7375 Genotypes, 17, 46, 67
with Y-STR profiling, 110 German Stain Commission, 101
Index 169
Gesah case, 134135 Insects, 43

Gill, Peter, 6, 86, 88, 144 Intelligence-led policing, 2324
Gloves, 149 International Society of Forensic Genetics
Goldman v. Simpson, 8 (ISFG), 70, 92
Gosling, Ray, 2 Interpretation, 66, 71
Grim Sleeper case, 127128 analytical threshold, 6061, 96
expert disagreement, 143
H with low level DNA, 85, 86, 87
with mixtures, 96, 98, 113, 138, 142143
Hair objectivity, 14, 141143
color prediction, 44 prior to reference sample comparison,
degradation, 19 97, 98, 142
from domestic animals, 128129 subjectivity, 66, 141143
evidence, 40 uninterpretable profiles, 101
mitochondrial DNA profiling, 120, 121,
122 J
reference samples, 44
Haplotypes, 105, 108, 109 Jackson case, 140141
cited in report, 149 Jama case, 2021, 135, 141142, 144, 149
frequency estimates, 70, 110111, 111113 Japan case, 107108
meaning of Y-STR match, 111113 Jefferson, Thomas, 108
surveying method, 112 Jeffreys, Alec, 5, 6, 45
HardyWeinberg equilibrium, 68 Junk DNA, 4
Hearsay, 9 Juries
Heilbronn phantom case, 93, 136137 considerations of, 72, 73
Heredity, 1, 17, 77 and CSI effect, 30
Heteroplasmy, 120, 121 decisions by, 8, 9, 27, 42, 100
Heterozygote, 61, 97 information conveyed to, 30, 84, 96, 145
Heterozygous, 1, 17
HIrisplex test, 44 K
Homozygous, 1, 17
Human genome project, 34 Kercher murder case, 2728, 93, 135, 138
Human hairs, see Hair Knox case, 2728, 91
Human remains identification, 24, 117, 118, Krone case, 1415
119
Hypotheses, 13, 36, 67, 133 L
I Laboratories; see also Reports

accreditation, 55, 61, 153
Identical twins case, 21 audits, 132
Identity, 14, 17, 71 background DNA levels, 135136
Inbreeding coefficient, 68, 78 contamination in, 133, 134, 136
Inclusion error, 8
combined probability, 69 process reviews, 61
standards for, 142 protocols, 153
as used in forensic statements, 7375 quality control, 132133, 140
with Y-STR profiling, 110 records, 151
Inconclusive findings, 51, 74, 76, 143 sterility versus DNA-free materials, 136,
Inheritance, 12, 17 137
Inhibition of DNA, 59, 81, 82, 83 test validation, 55, 132
Innocence Project, 7, 10, 13, 75, 109, 121 Ladders, 54
170 Index
Law of Independent Assortment, 2 detection protocols, 85

Law of Segregation, 1 enhancement techniques, 84, 90
Legal cases experiment using, 8890
absence of evidence, 15 interpretation of profiles, 85, 86, 88
collection practices, 8, 2728, 84 legal cases, 2728, 84, 138
compromised evidence, 8, 2728, 84 mixtures, 99
confessions, 6, 15, 107108 and partial profiles, 81, 82
contamination, 2021, 132133, 134, 135, reliability of results, 82, 9092
136137 statistical models, 9192
DNA from maggots, 43, 44 stochastic effects, 8586, 87, 88
DNA stability, 1819 Low template DNA, see Low level DNA
domestic animal hair, 128
errors, 140141 M
familial DNA searching, 126127,
127128 Maggots, 15, 43, 44
first murder solved by DNA, 6 Masking, 66, 85, 88, 92, 97
involving mixtures, 75, 9596, 100 Mass disasters, 24
low levels, 2728, 84 Mass graves, 117, 119
mass screening, 6, 108109 Mass screenings, 6, 108109
mitochondrial DNA, 120, 121 Matches, 20; see also Likelihood ratios;
partial profiles, 2728, 84 Probabilities
precedent setting, 128 binary, 71
questionable science, 1415 coincidental probability, 111112
relevant evidence, 2728, 29, 37, 3839 in databases, 18, 22, 23, 71, 127
statistics, 21 no-match systems, 71
time of deposition, 2627 partial, 126
with touch DNA, 4243 reporting, 111113
use of database to solve cold case, 42 subjectivity in, 142
warning signs, 131, 149150 with Y-STR profiling, 109, 110
with wearer DNA, 4142 Maternal inheritance, 117, 120
Legal fallacies, 7173 Mendel, Gregor, 1
Leskie case, 134 Mendelian inheritance, 12, 68, 77, 126
Likelihood ratios, 20 Mengele, Josef, 7
approach, 6668 MeowPlex kit, 129
basis for, 67 Messenger RNA (mRNA), 36
with mixtures, 68, 97, 99, 100, 113114 Microbial forensics, 129
versus probability of exclusion, 6970 Microsatellites, 6, 17
sampling uncertainty, 68, 76 Minifiler STR kit, 129
with stochastic effects, 88 Minisatellites, 5, 6
verbal descriptor scale, 76 Missing person cases, 18, 45
Linkage equilibrium, 68 Mitochondria, 119120
Locards Theorem, 25 Mitochondrial DNA (mtDNA) profiling, 4,
Loci, 17 17, 24, 40, 149
LoComatioN, 97 analysis of bone, 117119
Locus drop-out, 83, 85 contamination of samples, 124125
Low copy number (LCN), 84, 107; see also counting method, 111, 112, 123
Low level DNA disadvantages of, 120
Low level DNA discrimination of results, 120, 123
contamination, 9293 familial searching, 126128
cross-examination questions, 152 from hair, 120, 121
definitions, 82, 8485 and heteroplasmy, 120, 121
Index 171
inheritance, 117, 120 Paternity testing, 108

international database of, 123 Patrilocality, 110
legal cases involving, 69, 120, 121, 122, Pauling, Linus, 2
124125, 126128 PCR, see Polymerase chain reaction (PCR)
mixture profiles, 124, 125126 Peer reviews, 61, 132
regions, 121 People v. Castro, 7
relevant population, 122 People v. Simpson (1995), 8
statistical analysis, 122123 Personal belongings, 45
survival, 120 Personal protective equipment, 149
versus Y-STR profiling, 120 Peterson case, 122
Mixtures, see DNA mixtures Phantom of Heilbronn, 93, 136137
Multiplexes, 45, 46, 120 Phenotyping, 44
Mutations, 109, 110, 121 Pitchfork case, 6
Plant material, 129
N Plea bargains, 9, 143
Poland case, 108109
Nasal secretions, 34, 40 Polymerase chain reaction (PCR), 6, 18, 45
National Criminal Investigation Database number of cycles, 53, 82
(NCIDD), 22 purification post-PCR, 90
National Institutes of Health, 4 reduced volume, 90
Natural disaster victims, 24 Populations
Netherlands Forensic Institute, 139, 140 and ethnicity, 68, 77
New York Medical Examiners Office, 109 relevance, 77, 110111, 122
New York Supreme Court, 7 sample size, 68, 69, 110
Nichols, R.A., 69, 76 PowerPlex 16, 22
No-match systems, 71 Presumptive tests, 36, 59, 62
Nobel Prize, 2, 3 Pretrial hearings, 7, 8, 144
Novel techniques, 129 Pretrial reviews, 151
Nuclear DNA, 40 Privacy, 23, 127, 128
in cells, 17 Probabilities
profiling using, 17, 149 approach, 1921, 6667, 9192, 144
of Romanov family, 117 of coincidental matches, 111112
Nucleotides, 4 conditional, 68
of exclusion, 6970
O fallacy of the transposed conditional,
7173
Objectivity, 14, 141143 of guilt/innocence, 71, 72
O.J. Simpson trial, 8 Product rule, 68, 77
Omagh bombing, 84, 93 Propositions, 6768, 77, 98
Prosecution
P calling experts, 148
fallacy of, 7173
Paint evidence, 144 hypothesis, 67
Partial matches, 126 questioning DNA evidence, 148149
Partial profiles; see also Low level DNA responsibilities, 145
causative factors, 81 Puch-Solis, R., 65
cross-examination questions, 152 Pull-up, 57
electropherograms of, 58, 83
legal cases, 2728, 84, 141 Q
statistical value, 81
Paternal inheritance, 109, 110, 117, 118 Quality control, 132133, 140, 148
172 Index
R S
R. v. Deen, 1994, 72 Safeguards, 133
R. v. Doheny and Adams (1997), 8 Saliva, 34, 3940
Race, 67, 77 cross-examination questions for, 152
Random man not excluded (RMNE), 70 DNA degradation, 40
Random match probability (RMP), 6667, electropherogram of, 89
99 Sampling correction, 68, 69, 76
Rape cases, 7, 4243, 107, 108, 111 Sampling errors, 112
Bloodsworth case, 10 Science
Butler case, 109 discipline of, 13
Button case, 29, 133134, 150
exactness, 144
Dewey case, 18, 19
philosophy of, 144
DNA mixtures, 74, 75, 93, 94, 143
Scientific acceptance, 7
Durham case, 75
Scientific method, 13, 36, 132, 133
evidence collection, 39, 133
Scientific Working Group on DNA
evidence contamination, 131, 134, 135
fought on consent grounds, 25 Analysis Methods (SWGDAM), 78,
gang rape, 74, 113, 143 97, 98, 112, 137
Jama case, 2021, 135, 141142, 144, 149 Scientists, see Experts
penetration without DNA deposition, Scott case, 132133, 149
39 Screening tests, 36, 59, 94, 106
Pitchfork case, 6 Semen evidence, 29, 34, 3839; see also Rape
postcoital interval, 106, 107 cases
Scott case, 132133, 149 cross-examination questions for, 152
Thibodeaux case, 1415 differential extraction, 38, 107
Ware case, 120 location in context, 3839
Waring case, 7374 masked stains, 36
Rarity, 20, 71 sperm survival, 38
Reagent blanks, 51, 52 testing all available evidence, 150
Recessive genes, 1 Serology tests, 5, 16, 33
Reference samples, 4445, 135 Sex chromosomes, 2, 129
and bias, 98, 143 Sex markers, 17
comparison to crime sample, 66, 98 SGM Plus system, 47
Relative fluorescent units (RFUs), 54, 99 Shedders, 43
Relatives, 21, 68, 7778
Short tandem repeats (STR), 1718, 20,
Reliability, 82, 9092
4546; see also Autosomal STR
Repeat offenders, 23
profiling; Y-STR profiling
Reports; see also Case files
frequency tables, 153
declared contributors, 7576
markers, 17
differing hypotheses, 133
inclusion and exclusion, 7375 primers, 53
terminology, 76, 84, 112, 113 Siblings, 7778
testing rationale, 133 Simpson case, 8, 70, 133
Restriction fragment length polymorphism Size bias, 76
(RFLP), 5, 6, 45 Skin cells, 34, 40, 43
Retesting, 143144 Smith case, 42
RNA, 36 Snaggle tooth killer, 1415
Romanov family, 117, 118, 121 Software, 54, 60, 70, 95, 153
Royal Statistical Society, 65 Spermatozoa, see Semen evidence
Russian royal family, 117, 118 State v. Woodall (1987), 7
Index 173
Statistics, 1921; see also Likelihood ratios; Thibodeaux case, 1415

Probabilities Thompson experiment, 8890
agreement on use of, 147 Touch DNA, 34, 138, 148, 151
assumptions, 71, 112, 144 case involving, 4243
Bayesian approach, 6869, 112 mixtures, 93
biological model, 91 Trace evidence, 25, 34; see also Touch DNA
combining autosomal and Y-STR, 114 Traits, 1, 2
confidence intervals, 112 Transcription errors, 140, 141
conservativeness, 92, 97, 112 Transfer
counting method, 111, 112, 123 case involving, 2627
exclusion percentage, 123 between cases, 132, 134
frequencies, 6869, 110111, 112, 113, 123 by crime scene personnel, 133
guides to, 65 between exhibits in same case, 138, 149
in mitochondrial DNA analysis, 122123 to firearms, 136
with mixtures, 97 modes of, 73
models, 9192, 97 to packaging, 138
2p rule, 92 during police searches, 136
product rule, 68, 77 primary, 26
random data, 142 secondary, 25, 26
reporting method, 123 solving cold cases, 19
sampling errors, 112 two-way, 27, 131, 150
short cut calculations, 92 unintended, 153
significance, 66, 69, 70, 85, 121 Trial preparation, 144
understanding, 65 TrueAllele, 97
weights, 110 Twins case, 21
of Y-STR profiling, 110113
Stochastic effects, 82, 8586, 88 U
accounting for, 90
in complex mixture, 101 Unbalanced peak heights, 88
electropherograms with, 87 Uncertainty, 76
Stochastic threshold, 47, 61, 90 United Kingdom (U.K.)
STR, see Short tandem repeats (STR) database, 22
Stutter, 88 legal cases in, 6, 126127
Subjectivity, 66, 141143 as pioneers of familial DNA searching,
Suboptimal profiles, 66, 82; see also Low 126
level DNA review of interpretation with low level
Suspect screening, 23 DNA, 85
Sutton case, 9596, 139 Unrestricted combinatorial method, 97
Sweat, 41 Urine, 43
U.S. 13-locus CODIS system, 101
T U.S. Department of Energy, 4
Tapp murders case, 134135

V
Targeted operations, 23
Taylor case, 37 Vaginal secretions, 4041
Technology, see DNA technology Vaginal swabs, 39, 93
Teeth, 118, 119, 125 Valentines Day murder, 126127
Terminology, 76, 84, 112, 113 Validation studies, 20, 61, 86, 132
Testimony, see Experts Verbal descriptors, 76
Texas Sharpshooter Fallacy, 142 Victim DNA, 15, 36, 47, 93, 150
174 Index
Vincent, Justice F., 20, 21, 135, 144 Y

Viruses, 129
Voir dire examination, 9 Y-chromosome, 109
Y Chromosome Haplotype Reference
Database (YHRD), 110
Y-STR profiling, 4, 17, 105106
W
benefits of, 106107, 108, 109
Ware case, 120 cases with, 107108, 108109
Waring case, 7374 coincidental match probability, 111112
Watson, James, 2, 3, 5 combining with autosomal STR, 114
Wearer DNA, 34, 4142 frequency estimates, 110111
Weir, Bruce, 76 mass screening through, 108109
Wellcome Trust, 4 match report, 111113
West Virginia Supreme Court, 7 mixture ratios, 113114
White case, 126127 versus mtDNA profiling, 120
Wilkins, Maurice, 2 number of male contributors, 107, 113
Williams case, 141 statistics of, 110113
Woodall case, 7 theory of, 109
Wrongful convictions, see Innocence Project weakness of, 109

Forensic Dna Evidence For Criminal Justice Professionals: Introduction To

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Forensic Dna Evidence For Criminal Justice Professionals: Introduction To

Uploaded by

Copyright:

Available Formats

Introduction to

Boca Raton London New York

CRC Press is an imprint of the

No claim to original U.S. Government works

International Standard Book Number-13: 978-1-4398-9910-6 (eBook - PDF)

Chapter 1 History of forensic DNA profiling in criminal

Chapter 2 Strengths and limitations of DNA profiling evidence...... 13

2.12 Relevant evidence?................................................................................. 27

Chapter 3 DNA profiling basics................................................................ 33

Chapter 4 Evidential value and statistics................................................. 65

Chapter 5 Partial profiles, low levels, and mixtures............................. 81

Chapter 6 Y-STR profiling........................................................................ 105

6.5 Number of male contributors to Y-STR profile.................................113

Chapter 7 Other DNA techniques including mitochondrial DNA.... 117

Chapter 8 Concerns and controversies................................................... 131

Chapter 9 DNA pointers for criminal justice professionals.............. 147

9.7 Suggested cross-examination questions........................................... 151

Appendix A: Glossary of terms used in reports and testimony.............. 155

Purpose of this book

knowledge to do just that and make DNA profiling less intimidating.

Scope and limitations

government to attend the American Academy of Forensic Sciences meet-

History of forensic DNA profiling

1.1Discovery of structure and importance

during fertilization, and the Law of Independent Assortment, by which

U.S. National Library of Medicine

1.1.3Human genome project

After 13 years of laborious and meticulous work, the Human Genome

1.2DNA and concept of individuality

1.3Alec Jeffreys and the worlds first

solved by DNA profiling (Case 1) was prosecuted in the English midlands

Professor Jeffreys first exploited DNA analysis to confirm the iden-

1.4Early criminal court challenges

required to determine whether a testing laboratorys methods meet sci-

1.5Changing the face of forensic science:

Strengths and limitations of

is the best known example of a distinguishing technique. DNA profiling

alleged expert. Krone was dubbed the snaggle tooth killer

The advent of DNA profiling has no doubt improved the recognition of

Unfortunately, a mystique surrounds the utility of DNA profiling. The

2.2Discrimination power of DNA profiling

2.3Genetic basis for DNA profiling

and functioning of a living organism. Nuclear DNA is the DNA found

2.4Stability of DNA profiling

is performed if a match is found. The probability statistic is based on the

See Chapters 4 and 5 for more detailed discussions of statistics.

Database matches are mainly through criminal offenses. However, a

2.8DNA intelligence-led policing

Mass screening of suspects (such as in Case 1, Chapter1*)

Mass screening or DNA dragnets triggered privacy concerns in the

a mathematical one based on the birthday problem, which is discussed

2.10DNA evidence in context

may not have deposited or transferred detectable DNA. Principles of trace

2.11Time of deposition: Transfer

Primary transfer occurs when DNA is transferred from a person to

The uncertainty as to which cup was used by the offender and

This case is illustrative of the prosecution inferring the association

The criminal justice professional should be alert to the possibility of other

2.14CSI effect and the notion of infallible

2.15Relationships of lawyers and scientists

DNA profiling basics

3.2Biological materials allowing DNA profiling

Specific separate confirmatory tests are required to confirm the pres-

1.1Discovery of structure and importance

1.1.3Human genome project

1.2DNA and concept of individuality

1.3Alec Jeffreys and the worlds first

1.4Early criminal court challenges

1.5Changing the face of forensic science:

2.2Discrimination power of DNA profiling

2.3Genetic basis for DNA profiling

2.4Stability of DNA profiling

2.8DNA intelligence-led policing

2.10DNA evidence in context

2.11Time of deposition: Transfer

2.14CSI effect and the notion of infallible

2.15Relationships of lawyers and scientists

3.2Biological materials allowing DNA profiling

3.2.1Searching for DNA on exhibits

3.2.3Semen and spermatozoa

3.2.6Dandruff and skin

3.2.12Urine and feces

3.3.3Plucked hair samples

3.4Current profiling technique: Short

3.5Reading tables of alleles

3.6Obtaining DNA profiles

3.6.5Separation and detection

3.6.7Artifacts and other technical issues

3.7Time required to obtain DNA profiles

3.9Case documentation and review

4.2Interpreting DNA profiles

4.3Statistical approaches and obtaining

4.3.3Comparison of probability of exclusion and LR methods

4.3.4Identity and rarity

4.5Understanding reports: Common phrases

4.5.1Inclusion and exclusion

4.6Sampling correction and uncertainty

4.7Relevant population and impact

5.2Low level and suboptimal profiles