Andrey Doronichev’s Post

Welcome to 𝐓𝐡𝐞 𝐌𝐨𝐥𝐞𝐜𝐮𝐥𝐞 𝐀𝐈 𝐎𝐝𝐲𝐬𝐬𝐞𝐲. In our previous post we had a look at "𝐓𝐀𝐆𝐌𝐨𝐥: 𝐒𝐭𝐚𝐭𝐞-𝐨𝐟-𝐭𝐡𝐞-𝐚𝐫𝐭 𝐃𝐞 𝐍𝐨𝐯𝐨 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐒𝐦𝐚𝐥𝐥-𝐌𝐨𝐥𝐞𝐜𝐮𝐥𝐞 𝐋𝐢𝐠𝐚𝐧𝐝𝐬" We got to know that 𝐃𝐞 𝐧𝐨𝐯𝐨 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐥𝐢𝐠𝐚𝐧𝐝𝐬 that bind with known target proteins has the potential to solve one of the most challenging parts of the drug design pipeline. 𝐓𝐫𝐚𝐝𝐢𝐭𝐢𝐨𝐧𝐚𝐥 𝐝𝐫𝐮𝐠 𝐝𝐞𝐬𝐢𝐠𝐧 heavily depends on screening libraries to select suitable lead candidates. In this post we will have a more deeper look at 𝐓𝐀𝐆𝐌𝐨𝐥. Existing Generative AI techniques for de novo generation have yet to live up to their promise due to their generation of unrealistic molecules that fail many of the traditional metrics for successful drugs. Therefore, 𝐓𝐀𝐆𝐌𝐨𝐥, our proprietary solution for de novo generation, was specifically designed to alleviate these challenges and bring de novo generation to the point of practical applicability. We achieve this by using a novel optimisation technique that generates drugs while simultaneously optimising multiple properties. In particular, we consider : 1. Binding energy of the protein-ligand complex with the Vina Dock docking algorithm. 2. Drug Likeness (QED). 3. Synthetic Accessibility (SA). On generating binders for 100 target proteins (previously not seen by the model) covering a wide variety of diseases, 𝐓𝐀𝐆𝐌𝐨𝐥 achieves strong results across the three metrics, as shown in Table 1. 𝐓𝐀𝐆𝐌𝐨𝐥 represents the best available model for de-novo ligand generation and is accepted at the [ICML] Int'l Conference on Machine Learning ML4LMS workshop, a premier event in the field of deep learning. #ai #molecule #machinelearning #drugdiscovery #pharmacy

🚀 Exciting News in Drug Development and AI! 🧠🔬 Thrilled to share the latest research breakthrough in rodent behavior analysis for drug development! 🐭💊 At the time I was working for Boehringer Ingelheim, I collaborated with an incredible team of scientist (Heike Schauerte, Hanna Behnke, Thomas Wollmann, Philipp Schwarz, Georg Rast, Adrian Loy) to develop a cutting-edge human-in-the-loop system for rodent behavior classification. 🕰️💼 We've integrated AI for complex behavior prediction and active learning, propelling our research into uncharted territories, to overtake the state of the art in just one year! This would have been impossible without the support of the outstanding team of OPNME (Florian Montel, Markus Koester) 🚀 Key Highlights: 1️⃣ Active Learning: Tackling the labor-intensive nature of manual annotation, our system significantly streamlines the process. 2️⃣ Novelty Behavior Recognition: With AI at its core, our platform excels in predicting and identifying rare events, a game-changer in drug development research. 3️⃣ Multi-Class Classification: AI is capable of classifying a large class of behaviors on a video, highlighting in vivo drug effects in early research phase 🏆🚀 👥 Joining forces with brilliant minds in AI and General Pharmacology, our research demonstrates the power of collaboration in pushing boundaries. 🤝💡 Kudos to the team for their dedication and expertise. 🌐 Read the full article here https://lnkd.in/dw3XS-ue to dive deeper into our journey and discoveries! 📚✨ I specificaly thank Heike Schauerte and Georg Rast for their kindness all along the project! #AIResearch #DrugDevelopment #Collaboration #ArtificialIntelligence #Innovation #ScienceMatters

Thanks Dataiku for the feature in your #AIandUs series!

As scientists we have our own biases towards what's trending in #drugdiscovery. I am not sure the #privateequity #VC community always agrees. I am delighted that Jernej Godec, Jenna Hebert, James Kasuboski, Jasmina Marjanovic and Swetha Murali are joining us at #BostonDOT24 to share their thoughts on -Key drivers of innovation in drug discovery -Improvements in translating discoveries from the lab to the clinic -Impact of emerging areas like AI/machine learning, targeted degraders and molecular glues -Perspectives on current challenges and opportunities Why wait until then to get talking?! https://lnkd.in/e3ED9js2

The Impact of AI in Life Sciences. Watch now https://lnkd.in/emyVkJgg our first Vistatec Life Sciences – In Focus webinar hosted by Maria Mateos. Explore the upcoming shifts in the biopharma industry, unveiling key trends and the impact on clinical trials. #Biopharma2024 #Vistatec #LifeSciences #Trends

How Biotechs Are Plugging AI into Drug Discovery Problems P.S. If you like this post please 1) Like it (so I know it's useful) and 2) Repost and share it (so others can see it too)

Tempus AI $TEM went public last week, valuing the company at $6.7B on 2023 revenues of $531M. Despite promising growth in genomics and data analysis, the company faces significant challenges, especially in medical best practices, regulatory hurdles, and reimbursement shifts. Their AI product line currently contributes minimally to revenue, and the leadership team lacks crucial medical and regulatory expertise. For a detailed analysis of Tempus AI’s potential and the hurdles they face, read IPO Candy's latest blog post. #Biotech #Genomics #DataScience #HealthcareInnovation #AI #TempusAI #InvestmentAnalysis #HealthcareTech #IPO

Rediscover valuable insights into the intricate world of drug discovery through Lifebit's blog. Accompany us as we investigate the journey from identifying targets to the realm of Drug Discovery 2.0, where AI and ML technologies supercharge efficiency and efficacy. Our blog post provides a comprehensive overview of the drug development process, shedding light on the stringent standards and extensive clinical trials required to bring new drugs to market. As we adapt to the transition to Drug Discovery 2.0, powered by AI and ML, uncover how innovative technologies are reshaping drug development and offering practical solutions to longstanding challenges: https://hubs.li/Q02CcFG40 #DrugDiscovery #AIinHealthcare #Lifebit

🤖 Introducing Owkin K: A community of specialized AI agents. Their mission? To work with our data scientists on decoding the complexities of the immune system. We are excited to continue our ambition for the future of drug discovery, development and diagnostics. 🚀 Watch the video to find out how Owkin will use #AgenticAI in the future.

Personalized medicine is no longer a distant dream. Excited to share insights on the transformative impact of Generative AI (GenAI) and Large Language Models (LLMs) in revolutionizing life sciences. Navigating the complex terrain of AI in life sciences requires a delicate balance of innovation and ethics. Read more here: https://lnkd.in/esPkEt64 #LifeSciences #GenAI #LLMs #HealthTech #AI #DrugDiscovery #Innovation #Healthcare #Modak