iiinova — where your digital self keeps evolving

Reposted by Seva Viliuga

Gabriel Rocklin @grocklin.bsky.social · 19/11/2025

All data are available on Zenodo and we'd love to see what you can do with it! Cydney also made a nice notebook to run the predictive model (including DMS scan) in Colab! Preprint: biorxiv.org/content/10.1... Colab: colab.research.google.com/drive/1KNWvG... Data: docs.google.com/forms/d/e/1F...

biorxiv.org

Global Analysis of Aggregation Determinants in Small Protein Domains

Protein aggregation is an obstacle for engineering effective recombinant proteins for biotechnology and therapeutic applications. Predicting protein aggregation propensity remains challenging due to t...

1103

Seva Viliuga @proteinator.bsky.social · 06/01/2026

A proper way of commuting through piles of snow 🤣

020

Seva Viliuga @proteinator.bsky.social · 01/01/2026

Exchanging for an authentic Wienerschnitzel 🤠

010

Seva Viliuga @proteinator.bsky.social · 31/12/2025

Biggest achievement of 2025 😂😂 First ever self-made cake seems to be a success!!

1111

Reposted by Seva Viliuga

Kresten Lindorff-Larsen @lindorfflarsen.bsky.social · 18/12/2025

Are you interested in integrative structural biology, but feel a bit lost? Don't worry, we have you covered with our FEBS Advanced Course, Lost in Integration Vol. 2 — probing biomolecules with AI and experiments probingbiomolecules2026.febsevents.org network.febs.org/posts/integr...

0208

Seva Viliuga @proteinator.bsky.social · 18/12/2025

Folding models learn protein stability only implicitly. Without access to negative data, one can in principle make use of the folding free energy (dG) and the change in the free energy upon mutation (ddG). I believe simple aux losses could help for cases where a mutation is clearly disruptive.

000

Seva Viliuga @proteinator.bsky.social · 17/12/2025

Hi Thomas, that's indeed the case! However, this would also mean that one can reliably refold and score redesigned sequences only of those proteins whose structures were deposited post AF2 training cutoff date. This is a big limitation in our opinion!

100

Seva Viliuga @proteinator.bsky.social · 05/12/2025

Had a great time presenting our work on flexibility-conditioned protein structure design at @embl.org and right after our recent work on generating conformational ensembles of proteins at @euripsconf.bsky.social together with Nico. Follow along - we will be soon releasing more exciting things :)

050

Reposted by Seva Viliuga

Simon Olsson @smnlssn.bsky.social · 10/10/2025

New preprint out! We present "Transferable Generative Models Bridge Femtosecond to Nanosecond Time-Step Molecular Dynamics,"

1223

Reposted by Seva Viliuga

MLSB (in San Diego + Copenhagen) @workshopmlsb.bsky.social · 12/09/2025

You asked and we listened... @workshopmlsb.bsky.social is excited to be expanding to Copenhagen, DK at @euripsconf.bsky.social 🎉 Two workshops (San Diego & Copenhagen) will run concurrently to support broader attendance. You can indicate your location preference(s) in the submission portal💫

2106

Reposted by Seva Viliuga

Simon Olsson @smnlssn.bsky.social · 25/08/2025

Hope you all had a good summer. I'm very happy to announce the speaker line-up for the falls Chalmers AI4Science seminars! Hope to catch you all there!

082

Reposted by Seva Viliuga

Kresten Lindorff-Larsen @lindorfflarsen.bsky.social · 06/08/2025

If you’re interested in learning more about protein folding and misfolding, I’ve created a convenient reading list with a few essential papers: scholar.google.com/citations?us... scholar.google.com/citations?us...

47816

Seva Viliuga @proteinator.bsky.social · 17/07/2025

6/6 Amazing team work with great co-authors @leif-seute.bsky.social, Nicolas Wolf, Simon Wagner, @bioinfo.se, Jan Stühmer and @graeterlab.bsky.social Try out FliPS and BackFlip yourself, code and Google Colab tutorials are available on GitHub!

000

Seva Viliuga @proteinator.bsky.social · 17/07/2025

5/6 We introduce a framework in which we generate candidate protein structures conditioned on flexibility with FliPS and use BackFlip to select the candidates whose predicted flexibility profile best matches the target before running expensive MD simulations.

100

Seva Viliuga @proteinator.bsky.social · 17/07/2025

4/6 We also introduce BackFlip - an equivariant network that can accurately predict backbone flexibility as derived from MD simulations. Crucially, BackFlip infers flexibility solely from the backbone geometry without requiring evolutionary information, making it useful for de novo protein design.

110

Seva Viliuga @proteinator.bsky.social · 17/07/2025

3/6 In a series of experiments, we demonstrate that FliPS samples novel, realistic proteins with diverse secondary structure composition and a remarkable resemblance to custom target flexibility profiles, as verified in 300ns Molecular Dynamics (MD) simulations of the designed samples.

100

Seva Viliuga @proteinator.bsky.social · 17/07/2025

2/6 Our model FliPS is a conditional flow matching model for protein structure generation. FliPS receives a flexibility profile as conditional input feature and learns how to generate realistic protein structures while respecting target flexibilities.

100

Seva Viliuga @proteinator.bsky.social · 17/07/2025

1/6 One of the key features of functional proteins is their inherent structural flexibility. In our recent work at #ICML, we introduce flexibility to protein structure design! More in a thread below. Code / Tutorial: github.com/graeter-grou... Poster: W-109, Thu 17 Jul 11 a.m. PDT — 1:30 p.m. PDT

1228

Seva Viliuga @proteinator.bsky.social · 13/07/2025

First time transatlantic and such a view over Greenland 🤯 Wish there were more glaciers :/

000

Seva Viliuga @proteinator.bsky.social · 13/07/2025

Hahaha this is hilarious!! I’m now totally convinced I should play with it as well 🤣

010

Reposted by Seva Viliuga

The Align Foundation @alignbio.bsky.social · 08/07/2025

1/4 🚀 Announcing the 2025 Protein Engineering Tournament. This year’s challenge: design PETase enzymes, which degrade the type of plastic in bottles. Can AI-guided protein design help solve the climate crisis? Let’s find out! ⬇️ #AIforBiology #ClimateTech #ProteinEngineering #OpenScience

12320

Seva Viliuga @proteinator.bsky.social · 04/07/2025

This is not happening, right?

000

Reposted by Seva Viliuga

Simon Olsson @smnlssn.bsky.social · 03/07/2025

New pre-print from PhD student Hang Zou on warm-starting the variational quantum eigensolver using flows: Flow-VQE! Flow-VQE is parameter transfer on steroids: it learns how to solve a family of related problems, dramatically reducing the aggregate compute cost!

1132

Seva Viliuga @proteinator.bsky.social · 27/06/2025

"...we investigated the performance of ... inverse folding models when trained on different structural datasets and on held-out set of experimentally determined PDB structures." It would be interesting if they did it vice versa and ran inference on some FoldSeek or dark clusters of AFDB...

110

Reposted by Seva Viliuga

Matthew Cobb @matthewcobb.bsky.social · 26/06/2025

Just learned that Frank Stahl (of the Meselson and Stahl DNA replication experiment ("the most beautiful experiment in biology") died at the beginning of April, to no fanfare. Here's a lovely video of them reminiscing: www.youtube.com/watch?v=7-tn...

youtube.com

The Most Beautiful Experiment: Meselson and Stahl

YouTube video by Science Communication Lab

513772

Reposted by Seva Viliuga

Rocío Mercado Oropeza @rociomer.bsky.social · 23/06/2025

You can now check out a recording of Julija's excellent talk on the Chalmers AI4Science YouTube channel: www.youtube.com/watch?v=y_7a...

youtube.com

Dr. Julija Zavadlav --- Multiscale Modeling with Machine Learning Potentials

YouTube video by Chalmers AI4Science

052

Reposted by Seva Viliuga

Max Zhdanov @maxxxzdn.bsky.social · 21/06/2025

🤹 New blog post! I write about our recent work on using hierarchical trees to enable sparse attention over irregular data (point clouds, meshes) - Erwin Transformer, accepted to ICML 2025 blog: maxxxzdn.github.io/blog/erwin/ paper: arxiv.org/abs/2502.17019 Compressed version in the thread below:

1205

Seva Viliuga @proteinator.bsky.social · 15/06/2025

Attending conferences in Italy seems to be a great idea 🥸

020

Seva Viliuga @proteinator.bsky.social · 29/05/2025

I finally won a lottery and got the CA visa?! Something must be wrong.. Guess that means me + ICML 2025! 😀

010

Reposted by Seva Viliuga

Kresten Lindorff-Larsen @lindorfflarsen.bsky.social · 18/05/2025

Led by Fia Larsen in @rhp-lab.bsky.social, we mapped degrons in all ~1600 human transcription factors We find overlap between degrons and TADs, and show how negatively charged residues contribute to function by helping prevent degradation See preprint for much more: doi.org/10.1101/2025...

1567

Reposted by Seva Viliuga

BenjMurrell @benjmurrell.bsky.social · 15/05/2025

My lab, at Karolinska, in Stockholm, is looking for a PhD student with a computational/quantitative background to work on probabilistic/generative models of proteins (structure and sequence). The research will involve methods development, and applications in vaccine design.

22113

Reposted by Seva Viliuga

Gina El Nesr @ginaelnesr.bsky.social · 12/05/2025

someone said we need a meme

1296

Reposted by Seva Viliuga

Kresten Lindorff-Larsen @lindorfflarsen.bsky.social · 04/05/2025

While this paper looks interesting, let me just say (again) that (essentially all) NMR ensembles in the PDB are NOT thermodynamic ensembles or meant to represent these. They are "uncertainty ensembles" and using them to benchmark machine learning (or other) models of dynamics is not a good idea.

911922

Seva Viliuga @proteinator.bsky.social · 27/04/2025

Is there a stream of the talk or maybe there’s gonna be a recording?

140

Seva Viliuga @proteinator.bsky.social · 26/04/2025

Not sure it’s so trivial 😅 ENM assumes a harmonic potential, which by definition limits the exploration of states far from equilibrium state. If pico-(low)nanosecond dynamics is of interest, maybe it can be considered trivial; is it anyhow informative - is another question..

020

Reposted by Seva Viliuga

Sergey Ovchinnikov @sokrypton.org · 15/04/2025

looks like it works, I was able to tariff my numpy installation: pypi.org/project/tari...

0232

Seva Viliuga @proteinator.bsky.social · 21/04/2025

I think the more interesting msg is that one doesn’t gain a lot by scaling up the datasets contrasting to what still seems to be quite a consensus in the pLMs field… 😅

020

Seva Viliuga @proteinator.bsky.social · 09/04/2025

Not too good news given the quality of research you’re guys doing.. 😬😬 Fingers crossed this will be solved asap!

030

Seva Viliuga @proteinator.bsky.social · 05/04/2025

Join our efforts in biomolecular design at MPIP!

010

Reposted by Seva Viliuga

Gräter lab @graeterlab.bsky.social · 31/03/2025

A while online now already: Predict the barrier of a reaction without knowing the transition state and in the low data regime using Gaussian Process regr. Led by Evgeni Ulanov, with Ghulam, Kai and Pascal Friederich @ KIT. @mpip-mainz.mpg.de @hitsters.bsky.social pubs.rsc.org/en/content/a...

0154

Reposted by Seva Viliuga

Gabriel Rocklin @grocklin.bsky.social · 26/03/2025

Small proteins can be more complex than they look! We know proteins fluctuate between different conformations- but by how much? How does it vary from protein to protein? Can highly stable domains have low stability segments? @ajrferrari.bsky.social experimentally tested >5,000 domains to find out!

48736

Seva Viliuga @proteinator.bsky.social · 23/03/2025

Enjoyed a lot meeting all my old friends and listening to their great talks at Mosbacher Kolloquium this week! @leif-seute.bsky.social @maxfus.bsky.social

020

Reposted by Seva Viliuga

Gina El Nesr @ginaelnesr.bsky.social · 20/03/2025

We are so excited to see where these models and data go next! 📝Paper: biorxiv.org/content/10.1... 👩‍💻Repo: github.com/WaymentSteel... 👾Colab: colab.research.google.com/github/Wayme... Huge congrats to the co-authors @hkws.bsky.social, @ramith.fyi, Hasindu, @sokrypton.org, and doro!

biorxiv.org

143