Emanuel [email protected]

A new blog and technical report (bit.ly/4a6KL2G) from @IsomorphicLabs shows SOTA performance across hardest modelling tasks in drug discovery more than twice AF3 level on low similarity protein-ligand, antibody-antigen, and exceeding FEP+ accuracy on binding affinity

🚀 Announcing the Official @OpenTargets MCP! We've partnered with @AnthropicAI to provide AI with seamless access to high-quality genetics and target-discovery data. Open for the entire community to accelerate therapeutic development blog.opentargets.org/official-open-…

2025 FDA drug approvals nature.com/articles/d4157… The US FDA approved 46 new drugs in 2025, despite a tumultuous year at the regulatory agency. Read about the trends and highlights, and see the full list of approved therapeutics in our annual round up

Splendid day

Next week, our Director of Computational Biology, Melissa Davis, will take part in a panel discussion at the @UKBIOBANK Scientific Conference in London. Melissa will talk about how Isomorphic Labs is leveraging large-scale data to transform the future of drug discovery. If you can’t join Melissa at the event, register to watch online. Head to the comments for the link.

Honoured to be stepping into the role of President at @IsomorphicLabs. We've built an AI drug design engine that we can see day-to-day is leading to more and more breakthroughs in the design of better medicines for patients. But we're just getting started! Excited to work with @demishassabis and our team to drive our mission forward!

Curious about what it’s like to work at Isomorphic Labs’ Swiss location? In our latest blog post we're taking you inside IsoLabs' Lausanne office to learn about its history and meet some of the Isos based there. Since opening in 2023, Lausanne has grown into a close-knit hub where data engineers, ML researchers, computational biologists and software engineers work together to solve complex problems every day. Head to the comments to read the blog!

HI-FEVER: a Nextflow pipeline for the high-throughput discovery and annotation of endogenous viral elements. #EndogenousViralElements #EukaryoticHostGenomes #Genomics #Bioinformatics academic.oup.com/bioinformatics…

Excited to share AlphaDIA's publication in Nature Biotechnology! Our open-source DIA framework brings deep learning directly to raw MS data with feature-free processing, transfer learning for any PTM, and performance matching top tools. nature.com/articles/s4158…

I keep saying “drug discovery” but most of my audience does not understand what this means. Here’s a thread I worked on over the past week trying to distil down drug discovery - and why it matters in the age of AI. OPEN THE THREAD 🧵

Excited to give opening keynote at #BioTechXEurope in Basel on October 6, at 9 AM! I'll share about our progress @IsomorphicLabs and our framework for using ML to model biology across multiple scales, from atoms to humans, to enable rational in-silico drug design. See you there!

Our Chief Technology Officer, @SergeiIakhnin will be opening @BioTechX_ EU Congress with a keynote at 9am CEST on Monday, 6 October. AI is already helping us to understand disease like never before, but where are we going next? Sergei will explore a future where applying frontier AI unlocks faster breakthroughs and with an ambition to solve all disease. If you are planning to attend, don't miss it. Find out more about the event here: bit.ly/42EClLB

It was great to join our CEO, @demishassabis, and Head of Medicinal Drug Design @_rebecca_paul, on the latest episode of Bloomberg Tech (@technology) with @TomMackenzieTV, to share how we’re approaching building our AI drug design engine from first principles, our vision for the future of drug discovery, and what makes Iso different 1/2

Benchmarking gene embeddings from sequence, expression, network, and text models for functional prediction tasks biorxiv.org/content/10.110…

I feel very humbled to have been selected for TIME's 100 AI list for 2025! Thankful over the last 3 years to help to shape the @IsomorphicLabs drug design engine and to build the team leveraging it for drug discovery. #TIME100AI time.com/collections/ti…

The pan-cancer proteome atlas, a mass spectrometry-based landscape for discovering tumor biology, biomarkers, and therapeutic targets dlvr.it/TLyB5j

We’re excited to launch a powerful new capability at Ginkgo Datapoints: high-throughput, low-cost ADME profiling—optimized for machine learning from day one. Whether you’re prioritizing small molecules or training predictive models, early access to rich ADME data can dramatically accelerate and de-risk your pipeline. Now, you can unlock: ▪️ Scalable, automated workflows—the Ginkgo standard ▪️ Cost-effective pricing built for throughput ▪️ Rapid turnaround to keep pace with your discovery cycles ▪️ Model-ready datasets, enriched with structured metadata This is biology built for algorithms. No more fragmented data. No more waiting months. Just fast, clean, AI-compatible outputs delivered at scale. Let’s accelerate drug discovery together. For more information, check out our blog post or reach out directly: datapoints.ginkgo.bio/updates/smolde…

Will AI-designed proteins break biologics patents? If your IP strategy still relies on “≥80 % sequence identity,” it might be time for a rethink. Thread👇 Tools like ProteinMPNN from the Baker Lab, take an existing e.g. AbAg complex or enzyme and can replace substantial parts of the input with new residues, while maintaining or often improving the function of the original. There's many examples of proteins produced with < 50 % identity that function like the originals.

holy shit, it’s here! deepmind just released AlphaGenome. an AI model that reads 1 million bases of DNA and predicts how any mutation changes molecular function not just in single genes but across the entire regulatory genome. DNA is code, and you are software 1/

We're excited to welcome Dr. Ben Wolf, M.D., Ph.D., as our new Chief Medical Officer at Isomorphic Labs! Dr. Wolf brings nearly 20 years of biopharmaceutical experience to the team, and will be based in our new Cambridge, Massachusetts location as we advance therapeutic programs into the clinic using machine learning to transform drug design. Read more on our website: bit.ly/3G2p7QH