🤝Excited to partner with Tamarind Deniz Kavi to build towards an agentic AI protein designer. 🔁 Agentic protein optimization — Starting from a sequence, Biomni iteratively improves thermostability by orchestrating AlphaFold-2, ThermoMPNN, and reasoning over predictions and

NEW from me STAT: We've learned that DeepMind is going to fund CASP, the protein structure competition that brought Google DeepMind's #AlphaFold to prominence (and a Nobel Prize), as its NIH funding runs out. More: statnews.com/2025/07/21/cas…

HLE has recently become the benchmark to beat for frontier agents. We FutureHouse took a closer look at the chem and bio questions and found about 30% of them are likely invalid based on our analysis and third-party PhD evaluations. 1/7

The MLSB workshop will be in San Diego, CA (co-located with NeurIPS) this year for its 6th edition in December 🧬🔬 Stay tuned MLSB (in San Diego) as we share details about the stellar lineup of speakers, the official call for papers, and other announcements!🌟

I participated in the ADMET property subchallenge using the Tabular Prior-data Fitted Network (TabPFN) with minimal tuning. It didn't work well and was in the bottom half of the submissions. That's okay! See my GitHub repo for code and a short writeup: github.com/agitter/asap-p…

Do any retrosynthesis prediction algorithms work well enough to be useful in practice? I have a compound that was commercially available and no longer is. It isn't similar to anything in 4 synthesize-on-demand libraries. FutureHouse's Phoenix can't give a retrosynthetic pathway.

🧬Want to do a user study (Evolla vs. biologists using traditional bio tools like BLAST, Foldseek) on some proteins with known functions but not publically avaialble. 📊If you have such data, please contact. Looking for collaborators. Evolla: chat-protein.com

The OpenADMET teams are cranking away. John Chodera (he/him) James Fraser (A$AP J) and I started to formalize the organization and we will have a lot more exciting news in the coming months. For now, here is the hello world blog post: openadmet.org/community/blog…

Variant Effect Prediction with Reliability Estimation Across Priority Viruses 1. A new framework called EVEREST has been introduced to predict viral mutation effects with reliability estimation. This framework uses a curated benchmark of 45 viral deep mutational scanning

Out today in PLOS Comp Biol: journals.plos.org/ploscompbiol/a… Intrinsically disordered regions make variant prediction deceptively easy for benign changes but very hard for pathogenic ones. Our work shows why current tools struggle here, and why disorder-aware approaches are needed.

A new blog from OpenAI and Retro Bio describes a custom AI model (“GPT-4b micro”) that can design better Yamanaka factors, the proteins used to reprogram mature cells into induced pluripotent stem cells. In 2006, a scientist named Shinya Yamanaka discovered four proteins that

We just updated our manuscript "Understanding Language Model Scaling on Protein Fitness Prediction". Where we explained why larger pLMs don’t always perform better on mutation effect prediction. We extended beyond ESM2 to models like ESMC, ESM3, SaProt, and ESM-IF1. #ProteinLM