Protein Structure Database

The AlphaFold Protein Structure Database has 200 million structure predictions. Even so, running the monomer version is still necessary in special use-cases such as engineered proteins, chimeric proteins, novel proteins translated from newly recognized Open Reading Frames (ORFs), and more. For protein complexes, it is best to use the multimer version.

DeepMind and EMBL’s European Bioinformatics Institute (EMBL-EBI) have partnered to create AlphaFold DB to make these predictions freely available to the scientific community. The latest database release contains over 200 million entries, providing broad coverage of UniProt (the standard repository of protein sequences and annotations). AlphaFold DB provides individual downloads for the human proteome, the proteomes of 47 key research and global health organisms, and the manually curated Swiss-Prot subset of UniProt.

​