Truly useful autonomous chemistry agents are imaginable (but it's early) 🫣 AgentD turns GPT‑4o into a chemistry helper: 📑 pulls FASTAs ➜ 🧬 generates & refines SMILES ➜ 🧪 predicts 67 ADMET traits ➜ 🛠️ plugs into Boltz‑2 for 3D binding poses Two refinement loops lift

🧵[1/10] A busy week for K-Ras! We report three stories on covalent, mutant-selective targeting of oncogenic K-Ras mutations ACS Publications. Covalent chemistry has been key in drugging this once “undruggable” GTPase. Sotorasib and adagrasib are the only approved therapeutics.

New Online! Degrons: defining the rules of protein degradation bit.ly/40hyN0Q

The CZI Science workshop on AI Modeling Benchmarking and Evaluation for Biology (virtualcellmodels.cziscience.com/micro-pub/benc…) was well organized with representation from academic, industry, and independent groups. This is a useful summary even though it can't capture all the discussion and ideas.

reprogramming cells with transcription factors is our most expressive tool for engineering cell state traditionally, we found TFs by ~guesswork ICML Conference we're sharing NewLimit's SOTA AI models that can design reprogramming payloads by building on molecular foundation models

Gonna give it a try

great overview of DIMENSION portco NewLimit, pioneering tech in epigenetic reprogramming to ̶s̶l̶o̶w̶ reverse aging in pursuit of future life saving, life extending, medicines

🤝Excited to announce scverse X Biomni! Biomni now supports 10 core scverse packages—including scanpy, squidpy, scirpy, and pertpy—enabling researchers to perform single-cell, spatial, and perturbation analyses using natural language. biomni.stanford.edu/blog/scverse-b…

TRACeR story was rejected by the same Science editor before, which is ok, but the fact that 2/3 new papers did not mention or cite us made me very sad😔 I left academia, I don’t need Science paper, but I need our contribution to the field to be properly credited and recognized

New preprint alert! We developed a covalent small molecule that targets guanines in structured RNAs, guided by the FMN riboswitch structure. In collaboration with Rob Batey biorxiv.org/content/10.110… #RNA #DrugDiscovery #ChemicalBiology

Writing is thinking. Don't let AI do it all.

one of the great pleasures of life is stretching your brain and feeling it expand into a higher level of abstraction. first slowly, haltingly, uncertainly, then surely I wish I could bottle this up and give it to everybody

I think it is worth talking about the 1 Billion cell milestone. This is not because 1Bn is a large and round number (or at least not entirely)! These number of cells, with the rich perturbative type of data that Tahoe Therapeutics's platform generates, puts the number of training tokens

No driver mutations could be identified in 181 tumors. The authors do not know why. Perhaps, it's an algorithmic failure. Perhaps, it reflects non genetic contributions to carcinogenesis. What do you think ?

What’s the next PDB? Maybe it’s …the PDB. We just launched diffUSE, a structural biology initiative exploring protein motion & rethinking how we generate and use experimental data. Quick 🧵👇 diffuse.science

Featured on the cover of Science Magazine, BioEmu Microsoft Research, a deep learning system that rapidly generates diverse protein conformations for accurate insights into protein function. Illustration: N. Burgess/Science; Data: S. Lewis et al., science.org/toc/science/38….

RosettaFold 3 is out! biorxiv.org/content/10.110…

Today, we're releasing a software suite to train biomolecular neural networks called AtomWorks. We have already used it to train performant networks in our group -- we hope that others find it as useful as we do! Check it out at github.com/RosettaCommons…

this is one of the most remarkable technical blog posts I’ve ever read

From QM/MM to ML/MM: A New Era in Multiscale Modeling 1. This review article explores the evolution from Quantum Mechanics/Molecular Mechanics (QM/MM) to Machine Learning/Molecular Mechanics (ML/MM) methods, highlighting how ML potentials offer a faster alternative to QM while