Get the latest tech news
SimpleFold: Folding proteins is simpler than you think
Contribute to apple/ml-simplefold development by creating an account on GitHub.
SimpleFold does not rely on expensive modules like triangle attention or pair representation biases, and is trained via a generative flow-matching objective. SimpleFold challenges the reliance on complex domain-specific architectures designs in folding, highlighting an alternative yet important avenue of progress in protein structure prediction. SimpleFold is trained on joint datasets including experimental structures from PDB, as well as distilled predictions from AFDB SwissProt and AFESM.
Or read this on Hacker News
