This is a great conference for both free energy methods and force field development.

Interesting points about interdisciplinary research. Having people in the same space talking over a long period of time (months) is one way to address this. It tends to happen organically, though, and is hard to plan.

1. Kevin Gross and I just posted a new science-of-science preprint. This one explores the looming peer review crisis. As many of you know, it's becoming significantly more difficult for journal editors to find scholars willing to serve as peer reviewers for submitted manuscripts.

Folddisco finds similar (dis)continuous 3D motifs in large protein structure databases. Its efficient index enables fast uncharacterized active site annotation, protein conformational state analysis and PPI interface comparison. 1/9🧶🧬 📄 www.biorxiv.org/content/10.1... 🌐 search.foldseek.com/folddisco

A real shame that the future of CASP is in doubt. There is still a lot of progress to be assessed (complexes, nucleic acids, ligands, ensembles). CASP has also been a pioneer in how to run blind assessments. www.science.org/content/arti...

Excellent opportunity for a postdoc to join @jgreener64.bsky.social’s group at the LMB. You’ll help develop novel molecular simulation methods to optimise sequences for diverse design tasks in computational protein design. See: www.nature.com/naturecareer... Apply by 13 JUL #PostdocJobs #ScienceJobs

Very proud to send Filippo Bigi to Vancouver to give an oral presentation at @icmlconf.bsky.social about our investigation of the use of "dark-side forces" in atomistic simulations. The final version is here openreview.net/forum?id=OEl... and it's worth a read even if you already read the #preprint

Interested in a postdoc position combining computational protein design and molecular simulation? There is an open position in my group at the MRC-LMB in Cambridge, UK. www.nature.com/naturecareer... Feel free to message me with any questions.

Yuchi Guo, a former Masters project student of mine, interviewed me for a new podcast. We talked about science, open source software and what makes a good life. YouTube: www.youtube.com/watch?v=NZlE... Apple: podcasts.apple.com/us/podcast/m... Spotify: open.spotify.com/episode/1suM...

🚨 We're hiring a Bioinformatics Engineer! 🛠️ Develop #JuliaLang & Python tools to model structural and evolutionary features of IDPs 📅 Start: 1 Sept 2025 🎓 3+ years of higher education (Master’s or engineering diploma preferred) 👉 Apply now: emploi.cnrs.fr/Offres/CDD/U...

📢 New preprint: "A graph neural network charge model targeting accurate electrostatic properties of organic molecules" by @charlie-adams.bsky.social et al out now on @chemrxiv.bsky.social #compchem doi.org/10.26434/che...

It's an interesting phenomenon that some of the deepest questions about how life works have become what looks like impossibly obscure molecular biology stuck right at the back of Nature, which will never get covered by the science media. Like this. /1 www.nature.com/articles/s41...

The reversible molecular simulation paper is out in final form: www.pnas.org/doi/10.1073/.... Now including an example of fine-tuning MACE-OFF23 to fit the RDF of water. Thanks to the reviewers for useful comments.

The amazing Sofia Lövestam initiated the below project, when she became interested in the vault particles that we sometimes observe in #cryoEM images of brain-derived #amyloid filaments. www.biorxiv.org/content/10.1...

Glad to see the MACE-OFF paper out in final form. They show that MLIPs can be relevant to biomolecules, I'm sure there will be lots more work in this area in the coming years.

Benchmarking of an open, generic protocol shows decent performance for relative binding free energy calculations. These methods are gradually making MD an important part of the drug discovery pipeline.

Round 1 of our machine learning potential really wants oxygen atoms in water to cluster together 😂

New release of BioStructures.jl just out: github.com/BioJulia/Bio.... Featuring chi angle calculation, mmCIF secondary structure reading and bonded graph construction all thanks to Tim Holy.

Had a great time at the @feworkshop.bsky.social and @omsf.io conferences. The future of open molecular software and free energy methods is looking good. If anyone has any recommendations for Boston, MA for tomorrow then do let me know.

I just used the PyMod PyMOL plugin for the first time and am very impressed. It installed without a problem and let me model missing loops easily.

Had a great time at the Museu de les Ciències Príncipe Felipe in Valencia, Spain. Saw lots of kids engaged with science and the Chromosome Forest exhibition covering much of biology. I can just about forgive the artistic licence in the DNA model.

We are looking for a postdoctoral fellow to join us to explore telomerase regulation by histones. Please get in touch if this may interest you.

Is complexity bound to increase over time? Interesting article discussing some recent work.

This has been a long journey, but it's now out in final form. @kjamali.bsky.social academic.oup.com/bioinformati...

Our new review, led by @yuanqingwang.bsky.social, considers the future of force fields: molecular mechanics, machine learning, and the under-explored space between. pubs.aip.org/aip/apr/arti...

When you're training a force field and step 2 of the simulation looks like this.

Our study of Mad2 conformational changes is out in final form. A fun collaborative project to work that uses NMR and MD. @connyyu.bsky.social onlinelibrary.wiley.com/doi/full/10....

Adding hydrogens is often seen as an easy step during MD setup, but having tried a few options there are all sorts of edge cases and nuances. Reminds me of PDB parsing: easy in principle, effectively impossible to account for all real world cases.

Interesting work exploring an enzyme reaction with a machine learning potential. Looking forward to much more like this in the next few years.

Nice piece on society journals. All publishing scientists should be aware of the difference between a society journal and a for-profit journal, and should consider this when choosing where to submit.

Google Summer of Code projects are available this year to improve molecular simulation in Molly.jl. Feel free to ask me any questions. julialang.org/jsoc

Good news for open science.

The publishing culture in machine learning isn't perfect but I've always enjoyed how there are papers "only" on arXiv with more citations than entire for-profit journals.

new blogpost on protein binder design blog.booleanbiotech.com/protein-bind...

Interested to see where Achira goes, the team is great and bridging the gap between classical and machine learning force fields is the future.

A great article about the value of science, with a bonus dig at the academic publishing industry. Strong writing as ever from @mastroianni.bsky.social. www.experimental-history.com/p/funding-sc...

The universal approximation theorem isn't much use in the data limited regime. Thank you for coming to my TED talk.

On the International Day of Women and Girls in Science, we celebrate all female researchers who push the frontiers of knowledge! Introducing @noeliaferruz.bsky.social - a chemist @crg.eu aiming to design a digital mind capable of explaining its reasoning. 👉 aiproteindesign.com

A big new release of Molly.jl just dropped. Including improved performance, especially on GPU (from Giuseppe Gambini), and support for other GPU backends (from @leioslabs.bsky.social). github.com/JuliaMolSim/...

14/20 biological amino acids, without chiral preference, found on an asteroid. Cool stuff and a very ambitious project. www.nature.com/articles/s41...

We are looking for a group leader with exciting ideas to apply machine learning to biological challenges. Feel free to ask me any questions.

Ever wondered how the chemotherapeutic drug paclitaxel can target slow-dividing tumours? Check out our preprint Hale et al! @tom-hale.bsky.social found a novel non-mitotic mechanism of action for paclitaxel involving nuclear-cytoskeletal coupling via the LINC complex! www.biorxiv.org/content/10.1...

FYI the designed esmGFP is 58% identical to the most similar extant GFP sequence - hence "500 million years of evolution" - but nearly half of the novelty (the 42% unaligned part of the seq) is derived from one other protein. Originally caught by @btnaughton.bsky.social, source link in alt text

That concludes my list of favourite papers 2001-2023. I plan to do another one in a couple of decades... The list is collected for convenience at jgreener64.github.io/posts/favour.... (1/3)

Favourite paper of 2023 "Updated benchmarking of variant effect predictors using deep mutational scanning" by Livesey and Marsh. (1/4) www.embopress.org/doi/full/10....

Favourite paper of 2022 "End-to-end differentiable construction of molecular mechanics force fields" by Wang et al. (1/4) pubs.rsc.org/en/content/a...

Favourite paper of 2021 "Learning neural network potentials from experimental data via Differentiable Trajectory Reweighting" by Thaler and Zavadlav. (1/4) www.nature.com/articles/s41...

Favourite paper of 2020 "Supervised Contrastive Learning" by Khosla et al. (1/4) proceedings.neurips.cc//paper/2020/...

Favourite paper of 2019 "Learning Protein Structure with a Differentiable Simulator" by Ingraham et al. (1/4) openreview.net/forum?id=Byg...

Favourite paper of 2018 "Developing a molecular dynamics force field for both folded and disordered protein states" by Robustelli et al. (1/4) www.pnas.org/doi/10.1073/...