🐍 Want to publish your own Python package? Join pyOpenSci’s 2-hour online workshop Nov 6: ✔️ Build your first package fast ✔️ Learn the core files (pyproject.toml, docs, metadata) ✔️ Publish to PyPI with confidence 📅 Nov 6, 2025 | Online 🎟️ bit.ly/PythonPackaging

The emperor has no clothes

Nikolai Slavov@slavov-n.bsky.social · last yr.

An assessment of DNA language models concludes: ◼️ They do not offer compelling gains over baseline models Their performance is inconsistent and requires much more compute. arxiv.org/abs/2412.05430

This is the golden age of science informatics!

Martin Steinegger 🇺🇦@martinsteinegger.bsky.social · last yr.

Run BioEmu in Colab - just click "Runtime → Run all"! Our notebook uses ColabFold to generate MSAs, BioEmu to predict trajectories, and Foldseek to cluster conformations. Thanks @jjimenezluna.bsky.social for the help! 🌐 colab.research.google.com/github/sokry... 📄 www.biorxiv.org/content/10.1...

The new #RDKit blog post is an update of an older tutorial on displaying fingerprint bits. greglandrum.github.io/rdkit-blog/p...

Just started using Zarr - literally unbelievably good.

Zarr@zarr.dev · 2y ago

Your friendly neighborhood #chunked, #compressed, & #cloud-ready #n-dimensional array format is now verified as @zarr.dev on @bsky.app! Visit us at zarr.dev for more info!

Are you interested in getting involved in peer review of software? We need more editors! We are looking for folks with a background in #physics and also people with more stats / analytics background. Join our team. support the #opensource tools that drive #science 📢 Apply here: buff.ly/PKSpJ47

Currently making a slide deck in markdown and I've never felt so powerful

Cool application of a foundation model to drug design

@macinchem.bsky.social · last yr.

Looking at TabPFN with Polaris datasets macinchem.org/2025/02/06/l... #cheminformatics

polaris is an amazing tool with a similarly amazing development team! Try it out today if you work in Computer Aided Drug Discovery world

Cas@caswognum.nl · last yr.

Big milestone for @polarishub.io. We merged our first external contribution, thanks to @fastprop.bsky.social! 🎉 github.com/polaris-hub/... #opensource #benchmarking #ml #drugdiscovery

🚨 Extension Alert: The intermediate leaderboard submission deadline for the antiviral challenge has been extended to midnight on February 7th! Don't miss this chance to see how you stack up against other participants! Submit your results today: polarishub.io/competitions

This is obviously how scientific articles should be, and if we had an open ecosystem designed for users rather than profit, we would have had this years ago. Time to dump commercial publishers and their paywalls.

Mike Morrison@mikemorrison.bsky.social · 2y ago

Want to see a quick demo of the technology that might replace citations in scientific papers? Goodbye citations, hello embeds. I was just hoping for links in scientific papers, but this is even better!

I highly suggest all my followers working with Python to follow @pyopensci.org - great resources for packaging your code for improved reproducibility

Thank you to everyone who read and cited my papers last year - proud to say that I received 10 citations in 2024! Double digits for the first time🎉

It is roughly 10 lines of code to go from 1 GPU to N GPUs with pytorch DDP. Pointing this out so that everyone is aware and doesn't shy away from scaling their code

That C-C and C-N coupling are now viewed as boring (comparable to amide bond formation!!) is a testament to the work that Buchwald, Hartwig, and many others have done in the past 20 years.

An interesting read for all scientists publishing papers

Konrad Kording@kordinglab.bsky.social · 2y ago

There was this Tsai/Yuste/Shepard paper that meister famously eviscerated. markusmeister.com/2018/03/20/d... It seems that the authors never retracted the paper (or does ieee not mark retractions)? If not, does anyone know why not? @retractionwatch.com

LLMs aren't reasoning, and I tire of seeing articles acting like they are

📢 Calling all ML practitioners! Have you been waiting for an opportunity to prove how well your model performs on a blind, newly generated and consistent test set? You now have the unique opportunity to! Super excited to launch a first competition on @polarishub.io!

Polaris@polarishub.io · 2y ago

🦠 We’re excited to announce our first competition in partnership with @asapdiscovery.bsky.social and @omsf.io! Test your skills across three sub-challenges revolving around SARS-CoV-2 and MERS-CoV Mpro🧵 Full details: polarishub.io/competitions Blog: polarishub.io/blog/antivir...

I want to learn fortran - any recommendations on projects to try, toolchains to use, etc.?

My machine learning data splitting library astartes has just hit 64 stars on GitHub! ⭐ We built this to help rigorously quantify how well our models actually work - if you want to quantify how well you models extrapolates into new feature or target space, give it a look: github.com/JacksonBurns...

GitHub - JacksonBurns/astartes: Better Data Splits for Machine Learning

Better Data Splits for Machine Learning. Contribute to JacksonBurns/astartes development by creating an account on GitHub.

github.com

Graduate school is revising a paragraph seven times knowing full well that people will skim it just to get to the figures...