Congrats Diego!

Adapting ProteinMPNN for antibody design without retraining The neural network ProteinMPNN designs protein sequences capable of folding into predefined tertiary structures and quaternary assemblies. It has become widely used due to its high success rates when ...

"Adapting ProteinMPNN for antibody design without retraining"

Our several-years-old fix to ProteinMPNN's tendency to make weird antibody CDR sequences is finally preprinted

Three BioML starter packs now!

Pack 1: go.bsky.app/2VWBcCd
Pack 2: go.bsky.app/Bw84Hmc
Pack 3: go.bsky.app/NAKYUok

DM if you want to be included (or nominate people who should be!)

Seems like an easy heuristic to apply when working out where to avoid!

Two BioML starter packs now:

Pack 1: go.bsky.app/2VWBcCd
Pack 2: go.bsky.app/Bw84Hmc

DM if you want to be included (or nominate people who should be!)

Interesting, but there are some points to note. Embedding compression is not usually how pLMs are used for variant effect prediction- see e.g ESM1v. EMS1v model was optimised for DMS prediction, so no surprise it does well. Scaling should consider data and compute, as in some earlier work this year.

The protein design & informatics group at GSK is hiring in switzerland - looking for experience w/ ML (LLMs, GNNs, etc) and/or classical computational structural biology techniques

A overview blog on the good things being done in GSK AIML's Sequence Learning team
gsk.ai/blogs/gsk-s-...

Reposting with tags 🧬🧶

Hi all, The #livingphysics feed is live! If you are interested in the physics of life or #livingmatter, please save and pin the feed. If you do, please also "like" it as it is how I keep track of interest. 🦫🦋 #evosky 🧪 🧬🧶 🔬 🐋🌱
bsky.app/profile/did:...

Now comes the hard part of figuring out what all these damn things do