Considering presenting the 'Lab in the loop paper' at journal club next Wednesday, and showing how AI will take over all the PI's jobs  and eliminating all possibility of getting a return offer hmm

wonder if it's harder to engineer antibodies instead of antigens or the other way around and why

🤯Three (!!!) new papers today in Science Magazine on the application of generative AI for the de novo design of peptide/HLA binding molecules! Completely unique 3D structure and binding mode compared with natural TCRs and TCR mimics! Links to papers 👇

GPT-5 keeps resisting queries on chemokines involved in driving metastasis. Bearish. Time to go back to Claude.

I get the quandary that AI labs are in regarding perceived biosecurity risk, but having to jailbreak models to answer benign molecular biology questions is tiresome

evening run! if we could crank down the humidity that’d be fantastic

We have a new collection of protein structure generative models which we call Protpardelle-1c. It builds on the original Protpardelle and is tailored for conditional generation: motif scaffolding and binder generation.

it is time to build

How to get kicked out of a wet lab as a computational researcher: >get introduced to the concept of in vivo testing >see how much of a pain in the neck mice can be >get virtual-cell pilled >go up to your PI and ask him why not 'just' build a virtual mouse model >fired

the urge to channel my inner steve jobs and build an Apple-styled local GUI for protein modelling that is ✨aesthetic ✨

Great blogpost by Nick Boyd and Sam Guns from Escalante Bio (posted on the other site!) blog.escalante.bio/minibinder-des… They introduce mosaic for binder design, and show some impressive results, with 7/10 and 8/10 binders working for two targets! github.com/escalante-bio/…

low hit rate, but I wonder if this is a sign for the rise of LoRA adapters for epitope targeting, amidst other things

“Show me the mechanism of action.” “Uh. from the p-p-perturb seq? right here. Knocks down Gene X, and the cells shift into Cluster 7." “You used fkn UMAP again. FUCK. Zoom in.” “Sorry?” “Zoom. The fuck. In.” “…Okay.” “Do you see it?” “…See what?” “He doesn’t see it.

The robots are becoming self-aware. Here's their side of the story utkarsh.bearblog.dev/in-defense-of-…

Join the early team Coefficient Bio. We have an AI team that I’m incredibly proud of, and truly think is the most effective and exceptional technical team building in AI x bio. We're looking for people to: * Build and maintain robust AI systems to an exacting standard * Design

If anyone else is looking for teammates/ has ideas, feel free to shoot a dm!

Davis STA 35C needs this!!