The future of machine learning for small-molecule drug discovery will be driven by data | Nature Computational Science Perspectives - "we propose that a greater focus on the data for training and benchmarking these models is more likely to drive future improvement, and explore avenues for

Chai-1: Impressive results in our hands. biorxiv.org/content/10.110… #compchem

Enabling Open Machine Learning of DNA-Encoded Library Selections to Accelerate the Discovery of Small Molecule Protein Binders • This study presents a fully open framework for DNA-encoded library (DEL) screening, integrated with machine learning (ML) to discover small molecule

There is a lot of interest in applying ML to DEL screening data. Here’s our open model applied to a real-world dataset, that successfully predicted new single digit uM binders from Enamine Ltd 🇺🇦, and that is automated (no chemist decided what to purchase). doi.org/10.26434/chemr…

We've released a new version (v0.2.0) of the chai-lab Python package that makes it significantly easier to pass Multiple Sequence Alignments (MSAs) to the Chai-1 model. When provided, MSAs can improve the accuracy of predicted structures. github.com/chaidiscovery/…

🎙️ The Target 2035 podcast series is officially LIVE! In our pilot episode, meet our hosts - Milka Kostić, PhD, Rachel Harding and Opher Gileadi who share their personal motivations for joining the Target 2035 initiative. Tune in to the pilot episode now! podcasters.spotify.com/pod/show/targe… 🎧

If you like total synthesis and Wordle, check out Synthordle, a retrosynthetic analysis game created by Shenvi Lab associates Kevin Zong (GS 2) and Jonah Luo (HS sophomore!). rshenvi.github.io/synthordle/ Coming to Scripps Research Chemical Synthesis class in Winter ’25.

Smart data factory: Volunteer computing platform for active learning-driven molecular data acquisition • The Smart Data Factory (SDF) introduces a volunteer computing platform that harnesses the power of personal computers worldwide to accelerate quantum chemistry (DFT)

Practically significant method comparison protocols for machine learning in small molecule drug discovery. #machinelearning #compchem dx.doi.org/10.26434/chemr…

Does machine learning work on DEL datasets? Indeed it does. Sometimes. pubs.acs.org/doi/10.1021/ac…

For SPR data aficionados, all data and sensorgrams generated at Structural Genomics Consortium (SGC) for thousands of CACHE #1 and CACHE #2 compounds are now available at public.thesgc.org/hitconfirmatio… CACHE #3 data will be added next month... Conscience #compchem #CACHE

All participants to CACHE challenges are invited to join the overview paper. pubs.acs.org/doi/10.1021/ac… CACHE#1 learnings: -Workflows were diverse. -Most top methods included some ML. -Hits were weak: lots of room for improvement. Structural Genomics Consortium (SGC) Conscience #compchem

Ab initio characterization of protein molecular dynamics with AI2BMD nature Microsoft Research 1. Introducing AI2BMD: This new AI-powered biomolecular dynamics system allows for ab initio accuracy in protein simulations, greatly enhancing the exploration of protein conformations

Critical Assessment of Protein Engineering (CAPE): A Student Challenge on the Cloud ACS Synthetic Bio 1. The CAPE challenge is a groundbreaking student competition designed to advance protein engineering through a cloud-based, iterative design-build-test-learn (DBTL) cycle. Inspired by

Excited to share that the AlphaFold 3 model code and weights are now available for academic use. Looking forward to seeing what new research this unlocks and how the research community builds on AlphaFold 3 for scientific discoveries github.com/google-deepmin… 1/2

#AlphaFold3 is open-source for academic use now! See what our partner, Mohammed AlQuraishi, has to say about our upcoming OpenFold3 project and commercial availability. #BioAI #OpenSource #OpenFold nature.com/articles/d4158…

A new Science study presents “Evo”—a machine learning model capable of decoding and designing DNA, RNA, and protein sequences, from molecular to genome scale, with unparalleled accuracy. Evo’s ability to predict, generate, and engineer entire genomic sequences could change the

Save the date for the second annual Conscience Symposium on Open Drug Discovery! Join us in Montreal on April 7-8, 2025, for two days dedicated to advancing open and AI-driven drug discovery. Learn more: conscience.ca/event/save-the…

A well anticipated overview highlighting the first CACHE challenge, which brought together 23 research teams from 10 countries to collectively predict 1,955 compounds targeting LRRK2-a protein linked to #ParkinsonsDisease. pubs.acs.org/doi/10.1021/ac…

23 reposts on BlueSky vs 5 here. Just saying…😉