Move over AlphaFold? Here comes Meta AI
Laura L. Howes · C&EN Global Enterprise · 2022
Scientists at Meta, the parent company of Facebook and Instagram, have released the structures of over 600 million putative proteins in a database called the ESM Metagenomic Atlas . The structures are for proteins predicted to exist because of genetic data from large-scale metagenomic screens of soil, seawater, and other sources. The proteins themselves have yet to be isolated or identified using proteomic methods. The team describes the process it used to perform this feat in a paper, which has not been peer-reviewed (bioRxiv 2022, DOI: 10.1101/2022.07.20.500902 ). In July, artifical intelligence firm DeepMind announced that it had filled a database with predicted structures for almost all known proteins using its AlphaFold algorithm. The Meta AI algorithm (ESMFold) used to make the new protein models is not as accurate as AlphaFold, but it is quicker. Its speed is a result of how the tool predicts protein structures using a language