Sergey Ovchinnikov

ʙɪɴᴅᴄʀᴀꜰᴛ2 is out, and we're not waiting for the paper. The full code drops today, free for academic and industry use. We're releasing it early so you can start designing right now, and bring its full power to the current Adaptyv competition. github.com/PacesaLab/Bi...

BFVD v3 contains 5.8M viral protein structures, 16× more than v2; 75% high quality, filling a major gap in AFDB coverage. It fully covers 72.6% of reference proteomes and spans 72.7% of ICTV species. Great work by @eunbelivable.bsky.social et al. 📄 www.biorxiv.org/content/10.6... 🌐 bfvd.foldseek.com

Introducing highly experimental localfold.org Building on @martinsteinegger.bsky.social af2 webgpu port, @milot.bsky.social optimizations & jax ports of af3-like models by @marielle.bsky.social, Choonghwan Lee, Julia Buhmann. WARNING: runs directly on your 💻, may drain 🪫 & eat data📱 & overheat 🔥💻

ColabFold 1.6.3 is out! 2.5x faster, pip-installable, ipSAE+pDockQ2 scores. Thanks Choonghwan Lee, Marielle Russo, Gyuri Kim 🐍pip install colabfold[alphafold] CF2 Sneak Peak with AF3/Boltz/Protenix/ESMFold2… 🐍pip install "colabfold[alphafold3]@git+https://github.com/sokrypton/ColabFold@af3-preview"

Fold Spacer lets you fly through protein structures (Weekend project #2). It’s my first game: I originally set out to build a racer with structures as the tracks, but was a little too crazy. So it became this instead. You can upload your own structures. 🌐 martin-steinegger.github.io/Fold-Spacer/

Excited to share our now published paper @cp-cell.bsky.social highlighting advances in modeling the evolution of protein-protein interactions with MSA Pairformer. Big thanks to Zhidian Zhang, Olivia Tang, @eunbelivable.bsky.social @milot.bsky.social @martinsteinegger.bsky.social and @sokrypton.org!

🚀 Excited to share our new work: Absolute Stability Predictor! 📊: forms.gle/4ZnXZSnTBvay... Built the MGnify Stability Dataset (1.8M+ measurements) and developed stability prediction models, together with @grocklin.bsky.social @KotaroTsuboyama, @sokrypton.org and teams.

Yours truly is a proper scientist now! TL;DR: we used AI to redesign parts of essential cell machinery with only 19 canonical amino acids instead of 20. Why? Great thread by @harriswang.bsky.social provides more context and details. Let me talk a bit about the AI design part of this. 1/

1/ Excited to share our new paper in Science @science.org: “Toward life with a 19-amino acid alphabet through generative artificial intelligence design.” A great collab w/ Sergey's group @sokrypton.org at MIT @columbiasysbio.bsky.social science.org/doi/10.1126/... 🦠🧬🛠️🖥️💥

I'm excited to announce some major updates to our ProteinEBM paper with Chenxi Ou @sokrypton.org!

Our paper with @sokrypton.org using AlphaFold2 to predict small-molecule binding sites in proteins is now out in Nature Methods! 🧵 rdcu.be/e7SnX www.nature.com/articles/s41...

New preprint🚨 Imagine (re)designing a protein via inverse folding. AF2 predicts the designed sequence to a structure with pLDDT 94 & you get 1.8 Å RMSD to the input. Perfect design? What if I told u that the structure has 4 solvent-exposed Trp and 3 Pro where a Gly should be? Why to be wary🧵👇

As a bonus, here's a video of ProteinEBM folding up the fast-folder NTL9, rendered in stunning 2D by py2Dmol from @sokrypton.org! We hope models like ProteinEBM can serve as a step toward solving the "real" protein folding problem.

An energy-based model of protein conformational space can be used to predict structure from sequence, sample from the conformational landscape, rank structures, and predict mutation effects. @sokrypton.org www.biorxiv.org/content/10.6...

I'm super excited to announce the first preprint of my PhD, together with Chenxi Ou and @sokrypton.org! ML has revolutionized protein modeling, but crucial challenges remain. For example, we can't reliably predict complicated protein structures without MSAs, which limits what we can design.

An interesting study from @aidenkzj.bsky.social, @abulnaga.bsky.social and @sokrypton.org that builds on our Progres model to find pairs of proteins with circular permutations: www.biorxiv.org/content/10.1...

Thrilled to share that the final piece of my PhD work is now on bioRxiv! biorxiv.org/content/10.1... With support from @nvidia and the @NSF, we used AlphaFold to screen 1.6M+ protein pairs, revealing thousands of potential novel PPIs. All data can be viewed at predictomes.org/hp

Is 3D dragging you down? Wish you could instead use the 2D ColabFold representation for all your work? 🤓 Introducing: py2Dmol 🧬 (feedback, suggestions, requests are welcome)

Working on the protein-hunter-chai google colab notebook. 😈 @yehlincho.bsky.social

Will it bind? A little worried about all the "TTTTTTT" 🧐 But looks cool 😎

Thrilled to announce our new preprint, “Protein Hunter: Exploiting Structure Hallucination within Diffusion for Protein Design,” in collaboration with @Griffin, @GBhardwaj8 and @sokrypton.org 🧬Code and notebooks will be released by the end of this week. 🎧Golden- Kpop Demon Hunters

Looks like someone has already tried to replace me with an AI agent 🫣

Exciting to see our protein binder design pipeline BindCraft published in its final form in @Nature ! This has been an amazing collaborative effort with Lennart, Christian, @sokrypton.org, Bruno and many other amazing lab members and collaborators. www.nature.com/articles/s41...

Now that OpenCRISPR is in nature and rekindled the 'what's-a-novel-sequence' debate, I'm happy to share an app to check this, which I built for fun some time ago. fuerstlab.shinyapps.io/SeqNovelty/ quick 🧵

MMseqs2 v18 is out - SIMD FW/BW alignment (preprint soon!) - Sub. Mat. λ calculator by Eric Dawson - Faster ARM SW by Alexander Nesterovskiy - MSA-Pairformer’s proximity-based pairing for multimer prediction (www.biorxiv.org/content/10.1...; avail. in ColabFold API) 💾 github.com/soedinglab/M... & 🐍

Excited to share work with Zhidian Zhang, @milot.bsky.social, @martinsteinegger.bsky.social, and @sokrypton.org biorxiv.org/content/10.1... TLDR: We introduce MSA Pairformer, a 111M parameter protein language model that challenges the scaling paradigm in self-supervised protein language modeling🧵

looks cool but they should really have cited and compared to gLM and gLM2, which are very conceptually similar: www.nature.com/articles/s41... www.biorxiv.org/content/10.1... I'll leave a biorxiv comment for the authors. It's hard to find all prior literature but this one is kinda an oof.