RT @[email protected]

Today we’re sharing a deep-learning method for protein design called RoseTTAFold Diffusion. With minimal input, it turns prompts (“create a molecule that binds X”) into new proteins that fold and function in the lab. We’ve tested 100s already.

PDF here: https://www.bakerlab.org/2022/11/30/diffusion-model-for-protein-design/

🐦🔗: https://twitter.com/_JosephWatson/status/1598409457855496192

A diffusion model for protein design - Baker Lab

DALL-E uses a diffusion model to create breathtaking images. Will vaccines and medications soon be made the same way?

Baker Lab
If this approach is as robust as these early results suggest, the impact on the molecular life sciences will be large. Not too long ago, a movie scene in which the hero asks a machine to create a protein with a custom function deserved major eye rolling. But now it’s happening.😳
@davidrliu i would love to see this work with photoswitchable molecules! Interactions that happen only in the presence or absence of light, preferably red light using an efficient light absorber…