AlphaFold

E39542

AlphaFold is an artificial intelligence system developed by DeepMind that predicts protein 3D structures from amino acid sequences with unprecedented accuracy, revolutionizing structural biology.

Aliases (1)

Statements (50)
Predicate Object
instanceOf artificial intelligence system
protein structure prediction system
protein structure prediction system
achievement top performance in CASP13 protein structure prediction competition
top performance in CASP14 protein structure prediction competition
announced 2020
announcedAt CASP13
approach end-to-end differentiable learning
structure prediction from sequence alone
basedOn attention mechanisms
evolutionary information
multiple sequence alignments
transformer architecture
citation Jumper et al., Nature 2021
dataSource UniProt protein sequences
developer DeepMind
DeepMind
Google DeepMind
evaluationMetric GDT_TS
field bioinformatics
computational biology
structural biology
hostedOn EMBL-EBI AlphaFold Protein Structure Database
impact accelerated determination of protein structures
reduced need for experimental structure determination
input amino acid sequence
language Python
license open-source license
majorUpdate AlphaFold2
notableFor high accuracy in protein structure prediction
near-experimental accuracy in protein structure prediction
revolutionizing structural biology
organization Alphabet Inc.
output protein 3D structure
parentOrganization DeepMind
published 2021
relatedTo CASP (Critical Assessment of protein Structure Prediction)
Protein Data Bank
releaseDate 2018
repository GitHub
sponsor Alphabet Inc.
supports proteins from many organisms
task protein structure prediction
usedBy academic researchers
pharmaceutical companies
usedFor drug discovery
functional annotation of proteins
protein engineering
uses deep learning
neural networks

Referenced by (4)
Subject (surface form when different) Predicate
DeepMind
Demis Hassabis
knownFor
DeepMind
developed
AlphaFold ("AlphaFold2")
majorUpdate

Please wait…