Don't be shy to take on a little two-week side project. These five months will be the most precious three years of your academic journey.

AlphaFold as a prior: experimental structure determination conditioned on a pretrained neural network Nature Methods - ROCKET improves experimental structure elucidation by integrating implicit structural knowledge from OpenFold, a trainable reimplementation of AlphaFold2, with X-ray...

ROCKET 🚀 inference-time optimization of AlphaFold to fit structural data is published! rdcu.be/fa9YH
Since our preprint, we’ve pushed it to regimes where other methods break: low resolution, weak signal, real experimental edge cases. Here’s what we learned: 1/15

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

New OpenFold3 preview out! (OF3p2)

It closes the gap to AlphaFold3 for most modalities.

Most critically, we're releasing everything, including training sets & configs, making OF3p2 the only current AF3-based model that is functionally trainable & reproducible from scratch🧵1/9

On the nature of the earliest known lifeforms Microfossils reported from Archaean BIFs most likely were liposome-like protocells, which had evolved intracellular mechanisms for energy conservation but not for regulating cell morphology and replication.

On the nature of the earliest known lifeforms

Five years ago, we released FLIP. The core question was: can ML models for protein fitness prediction generalize in the ways that actually matter for protein engineering, i.e. low data, extrapolation to more mutations, out-of-distribution sequences?

New pre-print from the lab on scaling transferable implicit transfer operators to protein dynamics. Collaboration with @olewinther.bsky.social lead by @panosantoniadis.bsky.social.

Introducing The Structural History of Eukarya (SHE): The first proteome-scale phylogeny constructed entirely from 3D structure.
We computed 300 trillion alignments across 1,542 species to map the tree of life. 🧵👇 (1/5)

These New AI Models Are Trained on Physics, Not Words, and They’re Driving Discovery These New AI Models Are Trained on Physics, Not Words, and They’re Driving Discovery on Simons Foundation

While most AI models are trained on text and images, the Polymathic AI collaboration has something different in mind: AI trained on #physics: www.simonsfoundation.org/2025/12/09/these-new-ai-... #science

New Preprint!! We show that binding entropy can be quantitatively predicted from crystallographic ensemble models, accounting for both protein conformational entropy and solvent entropy! www.biorxiv.org/content/10.6...

Multiple protein structure alignment at scale with FoldMason Protein structure is conserved beyond sequence, making multiple structural alignment (MSTA) essential for analyzing distantly related proteins. Computational prediction methods have vastly extended ou...

FoldMason is out now in @science.org. It generates accurate multiple structure alignments for thousands of protein structures in seconds. Great work by Cameron L. M. Gilchrist and @milot.bsky.social.
📄 www.science.org/doi/10.1126/...
🌐 search.foldseek.com/foldmason
💾 github.com/steineggerla...

Important perspective from Greenland.

UK Research and Innovation and its councils quietly quit X - Research Professional News Funding agency has “been using alternative platforms” amid ongoing controversies over social media company

UK Research and Innovation, plus its research councils, quietly quit X.

Via @robinbisson.bsky.social @sophieatrpn.bsky.social

www.researchprofessionalnews.com/rr-news-uk-r...

I found the secret Pret at Heathrow Terminal 5

Time for a ROK Espresso GC

I've had the pleasure of working on this in collaboration with my very sharp colleagues at Absci.

We look forward to feedback on all aspects of our design pipeline and preprint! [9/n, n=9]

This solves the false-negative problem completely! And you can use fancy asymmetric versions of ipTM @rolanddunbrack.bsky.social.

Interestingly we find that, of the 5 AFM-v2.3 models, model 2 is almost always the best on ab-ag complexes. We can cut runtime by 5x with little performance hit! [8/n]

It turns out that masking template AAs in "AF2Rank-Unmasked" may not be the optimal choice for ab-ag complexes. This is a much trickier modality for AFM—it is systematically less capable & confident there than on many PPIs seen in protein design papers.

So, we restore the template AAs! [7/n]

Unmasking AlphaFold to integrate experiments and predictions in multimeric complexes - Nature Communications Integrating AlphaFold (AF) predictions with experimental data is not straightforward. Here, authors introduce AF_unmasked, a tool to integrate AF with experimental information to predict large or chal...

The hack involves "unmasking" cross-chain template information. It's not how AFM was trained, but Mirabello et al showed that AFM can leverage the newly-unmasked information (www.nature.com/articles/s41...).

However, we find that "AF2Rank-Unmasked" suffers from similar issues as AF-IG. Why? [6/n]

AF-IG is serving the protein design community well, but it falls short on antibody-antigen complexes that are OOD to the training set—it throws away too many correct PDB structures.

Sth else? With a fun AF-Multimer hack, one can run AF2Rank on complexes (this is available on ColabDesign!) [5/n]

Improving de novo protein binder design with deep learning - Nature Communications Recently, a pipeline for the design of protein-binding proteins using only the structure of the target protein was reported. Here, the authors report that the incorporation of deep learning methods in...

A popular way to score/rank designed binders with AF is "AF Initial Guess" (AF-IG), introduced by the Baker lab. Here, the designed complex is fed as an initialization to AF's trunk (this is possible because of recycling).

Both AF-IG and AF2Rank are ways to provide AF with a structural hint. [4/n]

AF2Rank worked even better when template AA tokens were replaced with gap symbols. Otherwise, AF2 thinks you are giving it "the right answer" and, in the case of protein monomers, spits the decoy structure back to you over-confidently.

Why is this potentially useful to binder discrimination? [3/n]

Mohammed AlQuraishi on X: "Even ~2 years after AlphaFold2's announcement this paper (https://t.co/rNA0QWFIx9) remains my favorite in the post-AF2 realm. To be sure RFDiffusion is a strong contender and arguably has been more immediately useful, but I strongly believe this work will stand the test of time." / X Even ~2 years after AlphaFold2's announcement this paper (https://t.co/rNA0QWFIx9) remains my favorite in the post-AF2 realm. To be sure RFDiffusion is a strong contender and arguably has been more immediately useful, but I strongly believe this work will stand the test of time.

In 2022 (!), @jproney.bsky.social & @sokrypton.org showed (AF2Rank) that AF2's template track can be repurposed to perform SotA ranking of protein monomer decoys. You provide the decoy structure as a template (instead of the structure of another homologous protein) & look at confidence scores. [2/n]

This was fun work and a remarkable effort across the computational and wet-lab teams!

Strategies for in-silico filtering and ranking of antibody designs have been under-discussed in the literature, e.g. in most technical reports on antibody design that I've seen. Let's talk about these here! [1/n]

At Absci, we performed de novo antibody design campaigns against "zero-prior" epitopes—lacking structural data from antibody-antigen or protein-protein complexes.

Model architectures, training data curation, and scoring protocols are fully described.

Preprint: www.absci.com/wp-content/u...

Congressman Fine Introduces Greenland Annexation and Statehood Act to Strengthen U.S. National Security and Put Our Adversaries on Notice Today, Congressman Fine (FL-06) introduced the Greenland Annexation and Statehood Act, landmark legislation focused on securing America’s strategic national security interests in the Arctic and counte...

fine.house.gov/news/documen...

GERMAN PRESIDENT STEINMEIER:

“.. the United States has broken with the values that it helped to establish ..

“.. we have now moved beyond the stage where we can lament the lack of respect for international law or the erosion of the international order; we are far beyond that, I believe.”

Recently found out about @jvkersch.bsky.social's `tmtools` github.com/jvkersch/tmt.... It works nicely! Being able to pass user-defined sequence alignments is a nice (if simple) feature that is missing from OpenStructure's own `tmtools` @torstenschwede.bsky.social.

Black swan moment for energy production and storage?