Lab-in-the-loop therapeutic antibody design. At @PrescientDesign @genentech we have spent 3+ years reimagining drug discovery. We built a machine learning system to design and execute experiments. Here's how it works and what we can do 🧵. 1/

thanks to @kchonyc for helpful feedback, perspective, and hope for the future!.

If you, like me, felt frustrated after reviewing 6 papers for @NeurIPSConf and thought "there must be a better way" then this post is for you. Link 👇

congrats to @amyxlu on the publication of her work "Tokenized and continuous embedding compressions of protein sequence and structure" in @Patterns_CP ! . thanks to all co-authors! @wilson1yan @KevinKaichuang @GligorijevicV @kchonyc @pabbeel @RichBonneauNYC. Link below 👇

RT @anshulkundaje: @sara_mostafavi (@genentech ) & I (@Stanford) r excited to announce co-advised postdoc positions for candidates with dee….

RT @chaitjo: After a long hiatus, I've started blogging again!. My first post was a difficult one to write, because I don't want to keep re….

Thanks to @RowanSci @CorinWagen @AriWagen for sharing all their awesome findings!.

"Another Rowan user called this 'an AlphaFold moment' for the field. This really is a step change in how good molecular simulation can be.". This is exactly what we hoped for and intended when we set out to build OMol25. Super cool to see the excitement in the community.

The two most common questions I get via cold email are 1) what should I work on, and 2) how do I get a job doing research, engineering, etc.? . I wrote a new post summarizing my advice for people trying to enter Biology+ML, from undergrads to early career. 👇 (link below)

RT @pkoo562: The premier conference on Machine Learning for Computational Biology is Sep 9-10 at the NY Genome Center in NYC!. Submission d….

The future of AI/ML for science is multimodal and multiscale; credit to @OpenCatalyst and the entire OMol25 team for leading the way. 7/7.

Special shout out to @levine_ds for making it all happen on the biomolecules dataset. It's been immensely rewarding working with @JoshRackers, @levine_ds, and @bwood_m to make the definitive QM dataset for our field. 6/.

I'm excited to see what the scientific community does with these amazing resources, especially in drug discovery and biology. If you have questions, feedback, or just want to brainstorm, reach out to me!. 5/.

We've shared preliminary evaluations on intermolecular interactions and many other tasks, approaching chemical accuracy (~1 kcal/mol) for biomolecules and neutral organics. 4/.

With OMol25 and the accompanying foundation models, we can now model molecular interactions and much of the complexity of the molecular biology world at quantum mechanical accuracy. There are of course many limitations and much future work, which we discuss in the preprint. 3/.

First, the preprint, dataset, and models are here:. 2/.

Introducing Open Molecules 25, a foundational quantum chemistry dataset including >100M DFT calculations across 83M unique molecules, built with 6B core hours of compute!. What does this mean for drug discovery, biology, and BioML?. 1/

RT @kchonyc: what do you teach at the first-year graduate course on machine learning, in this era of LLM and large-scale compute? here's my….