Colby Souders on Machine Learning in Biologics Discovery at NextGen Biomed

Biologics are complex. Your AI strategy shouldn't be. Most AI models are great at predictions, but biological drug design is a different beast. At the recent Instytut Farmakologii im. Jerzego Maja Polskiej Akademii Nauk annual meeting, Krzysztof Rataj, PhD broke down the "Opportunities and Challenges" of using AI to design the next wave of biological therapies. The takeaway? Success isn't just about having the best algorithm; it's about knowing exactly where that algorithm meets its limits in a real-world project. Proud to have Krzysztof representing Ardigen at the intersection of science and technology! #biotech #lifesciences #AIinDrugDiscovery #biologics #computationalbiology #bioinformatics #AIinBiotech

🚀 Just attended an eye-opening webinar on “Artificial Intelligence Driven Natural Product-Based Drug Discovery”! From traditional natural compounds to cutting-edge AI technologies — the future of drug discovery is evolving faster than ever 💊🤖 The session by Dr. Partha Pratim Maiti was truly insightful, breaking down how Artificial Intelligence is revolutionizing pharmaceutical research and unlocking new possibilities in medicine. ✨ Key Takeaways: 🔹 AI is accelerating drug discovery like never before 🔹 Natural products + AI = Powerful combination for future medicines 🔹 Huge opportunities ahead for pharma students in AI-driven research This webinar gave me a fresh perspective on where the pharmaceutical industry is heading — and honestly, it’s exciting to be part of this journey! Grateful to Mandsaur University for organizing such a valuable session 🙌 #ArtificialIntelligence #DrugDiscovery #PharmaLife #FutureOfPharma #WebinarLearning #MandsaurUniversity #Innovation #StudentLife

New Post: Machine‑Learning Guided CRISPR‑Cas12a Base Editing for the Rapid Generation of Uniform IgG‑Secreting CHO Cell Lines with Minimal Genomic Instability - — ## Abstract Stable cell line generation remains a bottleneck in biopharmaceutical production, largely due to the stochastic nature of genomic integration, variable copy number, and unintended off‑target edits that compromise product quality and regulatory compliance. In this study we present an end‑to‑end platform that couples a transformer‑based deep learning model for CRISPR‑Cas12a target prediction \[…\]

Great to see DaltonTx recognised as one of Europe's leading AI drug discovery companies, transforming the future of medicine. AI will supercharge human expertise in drug discovery, reducing the cost of medicines and healthcare - one of our greatest global challenges. Founders Garry Pairaudeau, Anthony Bradley, Adrian Rossall and Charlotte Deane are building a platform to address that mission. Recognised as one of Europe's leading AI drug discovery companies in the latest European Dynamism report from redalpine and Sifted, Dalton is "building orchestration software linking AI models, lab results, and clinical data into a continuous loop — where the platform recommends the next experiment. This matters most for biologics, where antibodies can look promising in silico but fail in real systems." Dalton was founded on cutting-edge technology developed at the University of Oxford and decades of experience from companies including AstraZeneca and Exscientia, with experts in software engineering, machine learning, medicinal chemistry, and biologics engineering. I'm proud to have been a small part of that journey. Dalton represents the sort of AI we need to build, solving our most pressing problems, operating at global scale, and bringing the best of British and European technology to the world.

💡 How Recursion is helping to solve a key drug discovery bottleneck: ADMET prediction 📣 We recently announced the launch of the ADMET Network with Apheris – a federated data network designed for pharmaceutical companies to collaboratively train models for absorption, distribution, metabolism, excretion, and toxicity (ADMET) predictions without sharing proprietary data. Other founding companies include: Lundbeck, Orion Pharma, and Servier. Accurate ADMET modelling is one of the most challenging areas in drug discovery, accounting for an estimated 40–45% of clinical attrition. The ADMET Network enables pharma companies to collaboratively train robust ADMET models with an expanded applicability domain, better reflecting industrial discovery needs. By bringing together highly diverse proprietary ADMET datasets across members, it is assembling one of the largest distributed ADMET data foundations in the industry. “At Recursion, we apply AI end-to-end across drug discovery and development, with a deliberate focus on the bottlenecks in R&D where failure rates are highest,” says Najat Khan, PhD, CEO and President of Recursion. “ADMET remains one of the most persistent challenges in translating novel discoveries into successful medicines. By participating in the ADMET Network, we can materially improve the reliability of our early predictions by learning from a broader set of industry data—without compromising data ownership or IP. It’s a powerful example of how industry collaboration can accelerate innovation with real impact, helping deliver more medicines that matter.” 👉 Read about the ADMET Network in Genetic Engineering & Biotechnology News: https://lnkd.in/drZ693GG #admet #drugdiscovery #techbio #pharma #federatedlearning #AI

#ArtificialIntelligence is rapidly transforming how new therapies are discovered. We’re excited to see the continued progress of Diagen AI, a technology-biotech company developing generative AI tools for small molecule and peptide discovery, helping accelerate innovation across the pharmaceutical industry. As part of its next phase of growth, Aditya Tallapragada, President of AKT Health, has joined the Advisory Board of Diagen AI. https://lnkd.in/g8HdaVSy This collaboration reflects the growing convergence of #AI-driven drug discovery, clinical research expertise, and global #healthcare #innovation. The engagement will also support expansion across the UAE and Pacific Rim markets, where advances in #biotechnology, #AI, and #digitalhealth continue to accelerate. It’s an exciting moment to see #AIinnovation and real-world healthcare ecosystems coming together to shape the future of #drugdiscovery. #DrugDiscovery #Biotech #LifeSciences #HealthTech #DigitalHealth #AIinHealthcare #HealthcareInnovation #AKTHealth

🎉 We are delighted to announce that our research paper titled “Chem Crafter – AI Powered Drug Generator” has been successfully published. 📄✨ 🔬 This research explores the use of Generative AI in drug discovery, where advanced models are used to generate novel molecular structures and support the identification of potential drug candidates. Our work focuses on leveraging modern AI techniques to improve the efficiency of early-stage pharmaceutical research. 💊🧬 We are sincerely grateful to our guide Prof. GEETHA M.P for her constant guidance, encouragement, and support throughout the development of this research work. 👩💻 Authors: PRIYADHARSHINI M Sangamithra V SRIHARINI EN 🚀 This publication represents a meaningful milestone in our academic journey, and we are excited to continue exploring the potential of AI in healthcare and scientific innovation. #ResearchPublication #ArtificialIntelligence #DrugDiscovery #GenerativeAI #MachineLearning #StudentResearch

Revvity and Eli Lilly have announced a collaboration to expand access to AI drug discovery models through the Signals platform. The partnership aims to support federated learning and accelerate data-driven discovery across biotech research teams. Eli Lilly and Company https://lnkd.in/eSDwQgbw #ArtificialIntelligence #DrugDiscovery #Biotechnology #LifeSciences #MachineLearning

We’re excited to announce that Insilico Medicine is partnering with Liquid AI to build lightweight scientific foundation models for pharmaceutical research. Together, we’re moving beyond narrow, single-purpose systems toward generalist AI models that can operate entirely within secure, on-premise environments — enabling proprietary molecules, assays, and target data to stay private. Our first model, LFM2-2.6B-MMAI, delivers: • 98.8% success in multi-parameter molecular optimization (MuMO-Instruct) • Strong affinity prediction across 2.5M compounds / 689 targets • Robust chemical reasoning and 1-step retrosynthesis By combining Liquid AI’s efficient LFM architecture with Insilico’s MMAI Gym (1,000+ pharmaceutical benchmarks), we demonstrate that small, private models can achieve cloud-scale scientific performance. 🎥 Watch the conversation below featuring Alex Zhavoronkov and Ramin Hasani as they discuss what this means for the future of AI-driven drug discovery. For more info:  mmaigym@insilicomedicine.com. #AI #DrugDiscovery #FoundationModels #PharmaAI #LiquidAI #InsilicoMedicine

Nice. How significantly different will a Pharma ASIC be then say a diffusion model ASIC or a latent space or a GAN ASIC, if one were to exist ? I'm interested in knowing more about how you plan to build a general purpose Vs SLMs that are nicely orchestrated thru agentic workflow and get the job done. I thought the latter is the moat everyone wanted