Pretty sure you must have heard of the news!
BREAKING NEWS
— The Nobel Prize (@NobelPrize) October 9, 2024
The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Chemistry with one half to David Baker "for computational protein design" and the other half jointly to Demis Hassabis and John M. Jumper "for protein structure prediction." pic.twitter.com/gYrdFFcD4T
The two guys on the right developed the AlphaFold2 model, an ML model that predicts 3D protein structures from just the sequence. Here’s a deep look into the model:
A more recent version released is AlphaFold3 with a diffusion model used. Many open-source versions of its implementation have been produced with the best one being:
We're excited to introduce @ChaiDiscovery and release Chai-1, a foundation model for molecular structure prediction that performs at the state-of-the-art across a variety of drug discovery tasks
— Joshua Meier (@joshim5) September 9, 2024
We're releasing inference code, weights & a web interface: https://t.co/QmpbVO9Fhd pic.twitter.com/TU7xuOAaIF
And Modal Labs have implemented this model to run on their GPU architecture using simple, easy-to-use code.
Folding proteins on Modal with @chaidiscovery Chai-1.
— Joshua Meier (@joshim5) December 5, 2024
Comprehensive guide with working code samples from the @modal_labs team.https://t.co/IYSO7ppz4V
I was running into some issues with running the code and opened a PR and corrected it. Now, you can easily run the code yourself by following the instructions in
https://github.com/greenrace666/chai1modal
I’ve got a 1000 residue protein in under a minute!
Move around this structure, it’s 3D 👇 (Refresh Again)