Wikinventia — Atlas of discoveries and inventions · Digital Age

Atomic-precision protein structure prediction — AlphaFold / DeepMind

2020 AD · Transmission: Global
AIMethodBritish

In 2020 DeepMind presents AlphaFold 2 at the CASP14 competition (Critical Assessment of Protein Structure Prediction), solving with atomic precision a problem open since 1972: predicting a protein's three-dimensional structure from its amino-acid sequence. The protein-folding problem had resisted the best efforts of computational biology for 50 years — the gap between sequence and structure was the great unsolved problem of molecular biology. AlphaFold 2 solves it with a Transformer-based architecture and attention mechanisms that model spatial relationships between amino-acid residues, combining evolutionary information from multiple sequence alignments with structural geometry. The impact is immediate and unprecedented: in 2022 DeepMind publishes the predicted structures of more than 200 million proteins — practically the entire known proteome — freely, in the AlphaFold Protein Structure Database. It accelerates drug discovery, disease understanding, and protein engineering by decades. It is the first case in which an AI system solves a fundamental scientific problem of Nobel relevance: John Jumper and Demis Hassabis receive the 2024 Nobel Prize in Chemistry together with David Baker.

InstitutionDeepMind — London, United Kingdom
Historical regionUnited Kingdom
Primary sourceJumper, J. et al. — 'Highly accurate protein structure prediction with AlphaFold' (Nature, 596, 583–589, 2021). DOI: 10.1038/s41586-021-03819-2
Secondary sourceVaradi, M. et al. — 'AlphaFold Protein Structure Database: massively expanding the structural coverage of protein-sequence space with high-accuracy models' (Nucleic Acids Research, 2022); Nobel Prize — Chemistry 2024
Original languageEnglish
View this entry in the interactive atlas → View in graph →