
📣 MLCB + MLSB are joining forces under MLBio.org ! 🧬🤖 The first workshop will be held 2027, Join the MLBio mailing list to stay updated on calls for papers, deadlines and more info on the first event! groups.google.com/g/mlbio

Assistant Professor UAS at HES-SO Valais-Wallis 🇨🇭 🤗 HF Fellow. Working on AI, Protein Design and Open Science. Creator of bioicons.com

📣 MLCB + MLSB are joining forces under MLBio.org ! 🧬🤖 The first workshop will be held 2027, Join the MLBio mailing list to stay updated on calls for papers, deadlines and more info on the first event! groups.google.com/g/mlbio

Some studies earlier this year found that MSA pairing quality has no impact on structure prediction of co-evolving hetero-oligomeric complexes. I'm curious if co-folding models trained with the MSA pairing approach by @yoakiyama would do better
Dissecting the Black Box of AlphaFold in Protein-Protein Complex Assembly biorxiv.org/content/10.64898… この研究は、AlphaFoldがどのようにタンパク質複合体の構造を予測しているのかを理解することを目指しています。 特に、従来重要と考えられていた共進化情報ではなく、単体タンパク質の形状や界面の形の相補性が重要であるという視点で解析が行われています。 モデル内部の情報伝播を可視化することで、まず単体構造が決まり、その後に相互作用が推定されるという段階的な推論プロセスが示されています。 この理解は、protein designにおいて複合体を設計する際に、どの要素を重視すべきかを考える上で重要な示唆を与えます。

Excited to report the first de novo enzyme catalyzing two of the most energetically demanding reactions in biology—phosphomonoester and phosphodiester hydrolysis—with catalytic efficiencies comparable to natural enzymes! 🚀 desB was designed zero-shot with dEVA. No structure prediction, no pre-defined motif, no reaction-intermediates. 🧵 @StanfordBiosci @bioe_stanford @SLAClab @EPFL @hes_so @simonduerr

More info via this link or via DM jobup.ch/fr/emplois/detail/4…

We are hiring a PhD student 🎓 Work with me and Matthias Rüdt in our department on an exciting industry collaboration with Lonza combining high-throughput experiments and molecular modelling using AlphaFold and MolecularDynamics 📍 Sion, Valais, Switzerland

Today we're dropping the "beta" tag from Adaptyv, launching our new website and announcing our $8M seed round. When we started Adaptyv a few years ago, our core belief was: AI models for biology are only as good as the experimental data they're trained on and the hypotheses they can test in the real world. Now, after a year of working with many great partners, we’ve scaled our infrastructure to the point that we're now open to anyone who wants to use our platform! Overall, this year, over 30 companies started using Adaptyv to validate their protein designs - from some of the biggest pharmas to frontier AI labs to many, many techbio startups. We've run hundreds of experiments, tested well over 10,000 proteins this year and are generating the data that validates the best AI models currently in development.

Chai Discovery just released their new model Chai-2, validated on Adaptyv! Chai-2 enables generation of de novo proteins and epitope-specific antibodies using a multimodal generative architecture that integrates all-atom structure prediction and generative modeling. Chai tested many of their new protein designs using our high-throughput wet lab at Adaptyv to get high quality data to improve their AI models in as little as two weeks per round.
We’re excited to introduce Chai-2, a major breakthrough in molecular design. Chai-2 enables zero-shot antibody discovery in a 24-well plate, exceeding previous SOTA by >100x. Thread👇

In earlier work, we already analyzed how AllMetal3D given a structural input and no information about metal stoichiometry fares against co-folding methods given metal stoichiometry but no structure: doi.org/10.1101/2024.05.28.5…

We compared to a few other tools that perform either metal identity prediction or location prediction and found that is very important to include negative classes during training and also cluster metal ions by biological similarity.

We've made available a ChimeraX extension and a Webapp to easily run predictions. AllMetal3D is now also easily installable using pip. ChimeraX: cxtoolshed.rbvi.ucsf.edu/app… HuggingFace: huggingface.co/spaces/simond… Docs for Python Package: lcbc-epfl.github.io/allmetal…

Excited to share our new AllMetal3D model building on Metal3D. AllMetal3D predicts the location, identity and geometry of the major biologically relevant metal ions doi.org/10.1101/2025.02.05.6…

Thank you to everyone who came to MLSB this year!! We are collecting feedback in a short 2 minute survey (and more data means maybe next year we'll train a new model... or just get a bigger room 😄) forms.gle/WHFNfvxVh3QxuTXZA

Anyone working at MPG willing to share the license conditions ? I'd be curious if they have special terms. For the last contract they didn't. The info should be on this page: soli.mpdl.mpg.de/en/software…

Unbelievable how expensive BioRender is. @MPDL is paying 673 413 USD for a two year license for a software that isn't even compatible with open access publishing (even if BioRender claims otherwise). ted.europa.eu/en/notice/-/de…

Note that this is an Alpha release. It will soon be also possible to define modified residues, covalent modifications and view confidence. Ping me if you encounter problems.

Both the input component and the molecule output component are individually reusable. Simply pip install gradio_cofoldinginput or gradio_molecule3d.