Exciting symposium on AI and Biology at EMBO | EMBL in Heidelberg on 10-13 March 2026! Excellent lineup of invited speakers across various scales of biology! Deadline for abstract submission is coming up — Dec 2. 🔗 www.embl.org/about/info/c... #EESAIBio @EMBLEvents

I’m biased, but it’s well worth an hour or so of your time! #cryoEM #crystallography 🧪

OpenFold3-preview (OF3p) is out: a sneak peek of our AF3-based structure prediction model. Our aim for OF3 is full AF3-parity for every modality. We now believe we have a clear path towards this goal and are releasing OF3p to enable building in the OF3 ecosystem. More👇

ROCKET's webinar on cryo-EM+crystal data-guided protein structure prediction is now online 🚀: www.youtube.com/watch?v=_29C... Thank you to the many of you who attended and stayed overtime for more Q&A, and especially to @sbgrid.bsky.social for hosting us last week!

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

We're looking for a scientific program manager to help manage complex team projects in biomolecular modeling and machine learning that span academia and industry. If you're interested please apply here! opportunities.columbia.edu/cw/en-us/job...

First time at #MLCB! I'll be speaking tomorrow about my development of an ML predictor of domain–peptide interaction affinity to model proteome-scale signaling networks. It'll be livestreamed too.

Virtual seminar on ROCKET and inference-time optimization of AlphaFold predictions using experimental data. Alisia gives fantastic talks!

We'll cover our latest work on low resolution applications, a run-through of our codebase, and tutorials of how to run ROCKET on your own data. Join if you're interested!! 🚀

(1/7) Training biomolecular foundation models shouldn't be so hard. And open-source structure prediction is important. So today we're releasing two software packages: AtomWorks and RosettaFold3 (RF3) [https://www.biorxiv.org/content/10.1101/2025.08.14.670328v2](www.biorxiv.org/content/10.1...)

Announcing the Diffuse Project! We're unlocking protein dynamics through diffuse X-ray scattering - the overlooked signal that could revolutionize how we understand protein motion. seemay.substack.com/p/from-syste...

We are hiring a postdoc - come work with us (www.koremlab.science) at the intersection of #microbiome, data science, and women's health! Message or email me if interested. 🖥️ 🧬

In 1965, Margaret Dayhoff published the Atlas of Protein Sequence and Structure, which collated the 65 proteins whose amino acid sequences were then known. Inspired by that Atlas, today we are releasing the Dayhoff Atlas of protein sequence data and protein language models.

To say this is well-deserved would be the understatement of the century. Julia’s research vision is so clear-eyed, brilliant, and dare I say blindingly creative it gives me imposter syndrome (-:

I am beyond excited and honored to receive a BWF CASI! This amazing program will support my transition from postdoc to faculty as I continue to develop new modeling frameworks for elucidating and programming cellular behaviors.

Our work on deriving protein-protein (functional) associations across different human tissues from proteomics data is now published. There is quite a lot in this paper, so I will just touch on a few highlights: www.nature.com/articles/s41...

AFESM: a metagenomic guide through the protein structure universe! We clustered 821M structures (AFDB&ESMatlas) into 5.12M groups; revealing biome-specific groups, only 1 new fold even after AlphaFold2 re-prediction & many novel domain combos. 🧵 🌐 afesm.foldseek.com 📄 www.biorxiv.org/content/10.1...

Join us at Columbia on April 22 for the latest round of Postdoc Night Science NYC!

Protein dynamics was the first research to enchant me >10yrs ago, but I left in PhD bc I couldn't find big experimental data to evaluate models. Today w @ginaelnesr.bsky.social, I'm thrilled to share the big dynamics data I've been dreaming of, and the mdl we trained w them: Dyna-1. 📝: rb.gy/de5axp

Join Elizabeth Kellogg at our Machine Learning meeting. Submit your poster abstract by Feb. 27 for a chance to present alongside pioneers in microscopy, cryoEM, live cell imaging, and ultra-resolution technologies! #KSMLStructure #ML #AI @moalquraishi.bsky.social @possuhuanglab.bsky.social

Injecting arbitrary experimental data into AlphaFold has been a missing feature. With high-throughput crystallography yielding 1000s of structures/year (@covid_moonshot) & cryoET making 😍 but low-res in situ models, it's also become a necessary one for scaling up structural biology. Hence ROCKET👇!

Watch out for an upcoming bluetorial/tweetorial on this work from the amazing @alisiafadini.bsky.social! www.biorxiv.org/content/10.1...

Excited to share what I’ve been working on with @tkaraletsos.bsky.social to transform the field of drug discovery with atomistic foundation simulation models! I could not be more thrilled to be working with this dream team. achira.ai

Everything is chaos, but I wanted to share some awesome recent science from the lab that hints at where the future of biomolecular simulation is headed: Foundation simulation models that can be fine-tuned to experimental free energy data to produce systematically more accurate predictions.

If you're at #BPS2025 don't miss this! Julia gives amazing talks!

Enjoying #BPS2025! I'll present my development of a ML predictor of domain–peptide binding affinity to understand how affinity is optimized across the proteome for cell signaling. Finish out the meeting by coming to my talk tomorrow (Wed) @ 1pm!

We re-ran AF3 without templates, since we noticed it could use any template in the PDB, including the ground truth. We see the performance drops slightly in the lowest bins, but the gap to other methods still exists. We will update the preprint shortly!

It's long seemed that molecular biology is a natural home for ML interpretability research, given the maturity of human-constructed models of biological mechanisms—permitting direct comparison with their ML-derived counterparts—unlike vision and NLP. Our first foray below👇.

This might be the best paper on applying sparse autoencoders to protein language models. The authors identify how neural networks trained on amino acid sequences "discover" different features, some specific to individual protein families, other for substructures www.biorxiv.org/content/10.1...

It is indeed striking how much AF3-based methods struggle to generalize to novel protein-ligand interfaces. It is also interesting that there remains a meaningful delta between AF3 and other reproductions at the most difficult generalization task (~25% vs ~13% success rate, so roughly double).

Very cool work!!

Mark your 📅! Submit abstracts by Jan. 23rd, 11:59pm MST for our on Machine Learning / Imaging Biomolecules. Visit keysym.us/KSMLStructur... or keysym.us/KSImaging25. #KSMLStructure25 #KSImaging25 @moalquraishi.bsky.social @possuhuanglab.bsky.social @thevillalab.bsky.social @bcarra.bsky.social

Foldseek 10 with 4-27x (1-8 GPUs) faster search through MMseqs2-GPU. Faster ProstT5 protein search w/o structure prediction through multi-GPU/Apple Metal, new BFVD/BFMD databases and multimer clustering (preview). 💾 github.com/steineggerla... 📄 www.biorxiv.org/content/10.1... 🐍 available in bioconda

A framework for evaluating how well generative models of protein structure match the distribution of natural structures. @possuhuanglab.bsky.social www.biorxiv.org/content/10.1...

I'm organizing a Keystone symposium, along with Liz Kellogg and @possuhuanglab.bsky.social, on machine learning and macromolecules. Mar 23-26 in Keystone, Colorado. We have a great lineup and deadlines are coming up soon!

GET is finally published! - Paper: t.ly/iQct_ (new validations, dry and wet) - Model: t.ly/4jnUI (new tutorial on PBMC 10x Multiome data, and yes you can even fine-tune it on a Macbook) - Analysis package: t.ly/OqLAL - Demo: t.ly/rbFQB - Docker: t.ly/86n_i

Excited about L2G, led by Wenduo Cheng. We leverage LLMs to beat genomic FMs and strong supervised baselines on a wide range of benchmarks. L2G uses cross-modal transfer (rather than vanilla fine-tuning), and neural architecture search to learn a genomic-specific embedder model.

Excited to share my master's thesis work with Chris Kim, Nazim Bouatta, and @moalquraishi.bsky.social: QuickBind, a light-weight and interpretable molecular docking model (arxiv.org/abs/2410.16474).

Super excited to preprint our work on developing a Biomolecular Emulator (BioEmu): Scalable emulation of protein equilibrium ensembles with generative deep learning from @msftresearch.bsky.social ch AI for Science. www.biorxiv.org/content/10.1...

AF3 BEST METHOD followed by cluspro but also some conversion errors

Chai-1, the open source AF3 clone, is now available with an Apache 2 license github.com/chaidiscover...

Lazysky time: has anybody actually shown that structural self consistency correlates with successful expression on either a per design or per model basis?

Our Big Fantastic Virus Database (BFVD) is now published NAR! It contains protein structure predictions of major viral clades, enhanced by petabase-scale homology search and it's explorable on the web. 🌐 bfvd.foldseek.com 💾 bfvd.steineggerlab.workers.dev 📄 academic.oup.com/nar/advance-...

#GoodThingsComeToThoseWait Sorry it's taking us a while. Promise that the wait will be well worth it! #OpenFold3 Love all the great work that's already been done #HelixFold3 #Ligo #Chai1 #Protenix #Boltz1

I’m curious to see how all of the new AF3 mimics perform. 🧐 My lab’s been installing them on our servers, and faster inference and ease-of-use are key for us. Boltz-1 has an early lead, but nothing beats a good frozen pLM with a structure trunk! 😅 Bc accuracy to the PDB isn’t the best metric. 🤷🏾‍♂️

Really looking forward to welcoming & meeting the participants of our #EMBOStructBio course this morning! Many of our speakers/tutors are here: @moalquraishi.bsky.social @sykhalid.bsky.social @ezgikaraca.bsky.social @amjjbonvin.bsky.social @gchojnowski.bsky.social @drmelissagraewert.bsky.social

Thrilled to announce Boltz-1, the first open-source and commercially available model to achieve AlphaFold3-level accuracy on biomolecular structure prediction! An exciting collaboration with Jeremy, Saro, and an amazing team at MIT and Genesis Therapeutics. A thread!

developer.nvidia.com/blog/boost-a... Boost Alphafold2 Protein Structure Prediction with GPU-Accelerated MMseqs2 Nice improvements in speed