News

3D structure of human proteins

The European Molecular Biology Laboratory headed by Edith Heard, Professor at the Collège de France (Epigenetics and Cellular Memory Chair) recently announced an exceptional partnership with British artificial intelligence company DeepMind, to make the most complete and accurate database to date of protein structure model predictions from the human proteome freely available to the entire scientific community. These data were obtained using the AlphaFold algorithm developed by DeepMind to solve the problem of protein structure prediction, and whose source code is public.

Protein structures representing AlphaFold data. Image : AlphaFold. Design : Karen Arnott/EMBL-EBI.

This includes around 20,000 proteins expressed by the human genome. The data will be made available to the scientific community free of charge. The database and AI system provide structural biology researchers with powerful new tools for examining the three-dimensional structure of a protein, and offer a wealth of information that could unlock future advances and herald a new era in AI-assisted biology.

This work was published in the journal Nature on July 22, 2021.