Happy to share my first adventure in #TeamTomo in collaboration with the talented @inaharasimov.bsky.social. It was a fun experience experience with a lot of learning along the way.

Training a force field for proteins and small molecules from scratch Force fields for molecular dynamics are usually developed manually, limiting their transferability and making systematic exploration of functional forms challenging. We developed a graph neural networ...

Check out our pre-print, where we train a protein and small molecule force field from scratch with a graph neural network.

We show comparable performance to existing, manually-tuned force fields on a range of tasks including binding free energy prediction. (1/4)

arxiv.org/abs/2603.16770

We're excited to announce the beta-release of #RELION-5.1! 🥳 This release has important updates on our #amyloid processing pipeline, as described in this preprint: doi.org/10.64898/202... There are also some tweaks for #teamtomo.

#OpenSoftwareAcceleratesScience

Join #EMBOCryoEM to get a comprehensive foundation in advanced single-particle and cellular tomography techniques – we'll cover the entire workflow from sample preparation to data collection and image processing.

https://s.embl.org/cry26-01-bl
Apply by: 11 May 2026
@embo.org

Please spread the word: the Structural Studies Division @mrclmb.bsky.social is looking for a new tenure-track, independent group leader with an exciting plan in any area of Structural (Molecular & Cell) biology, in discovery biology and/or methods development. 🥳

mrc.tal.net/vx/mobile-0/...

Cow Tools!

We have lived alongside cows for nearly 10,000 years.
We breed them and exploit them

It is now, only now, that we have discovered THEY CAN USE TOOLS

Here I describe our study

(paper) www.sciencedirect.com/science/arti... in @currentbiology.bsky.social
with @auersperga.bsky.social

Excellent thread with links to tools whose development relied on the Chlamy dataset

The latest issue is now online www.cell.com/molecular-ce...

Final version is out! Our large-scale cryo-ET dataset 🔬 of Chlamydomonas reinhardtii 🦠 is now published in @cp-molcell.bsky.social

Huge collaborative effort! So glad to see the community already using it to develop new resources & tools.

Check it out here: shorturl.at/z4i4c
#CryoEM #CryoET

Postdoc opportunity with Michel Goedert and myself to work on #cryoEM/#cryoET structures of #amyloids. Feel free to email me with any questions you might have.

mrc.tal.net/vx/mobile-0/...

Join ML wizard Kiarash to develop BioML methods that work

An Asgard archaeon with internal membrane compartments

Brilliant study led by @fmacleod.bsky.social and Andriko von Kügelgen. Tight collaboration with @buzzbaum.bsky.social and lab. Congrats to all authors!

www.biorxiv.org/content/10.1...

Happy Halloween! 🎃

Here's a selection of some of our scariest gargoyles and grotesques 👀

#Halloween

Join our group to optimise Relion’s optimisation algorithms!

Google DeepMind is bringing AI to the next generation of fusion energy We’re announcing our research partnership with Commonwealth Fusion Systems (CFS) to bring clean, safe, limitless fusion energy closer to reality with our advanced AI systems. This partnership...

Very excited to be able to talk about something I've been working on for a while now - we're working with Commonwealth Fusion Systems, IMO the leading fusion startup in the world, to take our work on AI and tokamaks and make it work at the frontier of fusion energy. deepmind.google/discover/blo...

The final version of our multispectral microscopy manuscript is now online at Nature Photonics: www.nature.com/articles/s41...

The main additions are some further comparisons to yet more unmixing algorithms, in which we find that our algorithm is still the most accurate.

Sub-cellular chemical mapping in bacteria using correlated cryogenic electron and mass spectrometry imaging

Congrats Hannah Ochner and authors on this important paper! Strong collaboration with @kiranrpatil.bsky.social

www.biorxiv.org/cgi/content/...
@mrclmb.bsky.social @wellcometrust.bsky.social

Introducing DINOv3 🦕🦕🦕

A SotA-enabling vision foundation model, trained with pure self-supervised learning (SSL) at scale.
High quality dense features, combining unprecedented semantic and geometric scene understanding.

Three reasons why this matters👇

Cryo-EM structure of the vault from human brain reveals symmetry mismatch at its caps Lövestam and Scheres present a cryo-EM structure of vault particles, found during tau filament extraction from progressive supranuclear palsy (PSP) brain tissue. They reveal a symmetry shift at the va...

Sofia Lövestam's vault structure is now online in Structure @cellpress.bsky.social. 🥳

How beautifully explorative is #cryoEM! We never planned to do this. The vaults just came down with #amyloids from human brain tissue, and Sofia decided to have a closer look.

www.cell.com/structure/fu...

Targeted disruption of phage liquid crystalline droplets abolishes antibiotic tolerance of bacterial biofilms

led by @abultarafder.bsky.social

Great collaboration with @pearce-maths.bsky.social and @geiselbiofilm.bsky.social

www.biorxiv.org/cgi/content/...

How can we best use AI in science?

Myself and 9 other research fellows from @imperial-ix.bsky.social use AI methods in domains from plant biology (🌱) to neuroscience (🧠) and particle physics (🎇).

Together we suggest 10 simple rules @plos.org 🧵

doi.org/10.1371/jour...

Today, we release Franca, a new vision Foundation Model that matches and often outperforms DINOv2.
The data, the training code and the model weights are open-source.

This is the result of a close and fun collaboration
@valeoai.bsky.social (in France) and @funailab.bsky.social (in Franconia)🚀

openreview.net/forum?id=8tQ...

good link:
openreview.net/forum?id=8tQ...

Our entrance into protein design! Inverse folding steered by external sources of information, and for multiple conformations. By the amazing @kaiyi94.bsky.social and @kjamali.bsky.social!
Download our code and try it yourself. 🥳

Thanks! I did not try it on symmetric structures. However, preconditioning influences the speed of convergence to a minimum (good or not), rather than which minimum it converges to, so it's not clear to me whether it will avoid the problem you mention without running it.

Exciting study from @kiranrpatil.bsky.social and colleagues. Hannah Ochner from the group performed cryo FIB imaging.

Congrats!

“Achieving individual tomograms of cells where any features are at 10 Å resolution or better is well-beyond both current capabilities and reasonable predictions of future improvements.”
:(

Postdoctoral Scientist - Structural Studies - Dr Sjors Scheres - LMB 2640 - Cambridgeshire job with MRC Laboratory of Molecular Biology (LMB) | 12840844 Postdoctoral Scientist Starting Salary: £41,344 Fixed term: 3 years MRC Laboratory of Molecular Biology, Cambridge, UK We are looking for a postdoc...

We have a 3-year funded postdoc position to work on structures of #amyloid filaments. If you are an expert in protein biochemistry, structural biology or #cryoEM / #cryoET, this could be for you! Feel free to email me with any questions. Come join us! 🤩

www.nature.com/naturecareer...

We also present a solution to the problem, showing how we can efficiently estimate 2nd order information and largely mitigate the poor conditioning, opening the door for efficient, SGD-based high resolution cryo-EM methods.