# AlphaFold 2: predicting protein structures to near-experimental accuracy

Source: https://onco.cc/key-papers/paper-alphafold2-jumper-nature-2021/  
OnCo record `paper-alphafold2-jumper-nature-2021` (Key paper). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

DeepMind's neural network predicted protein structures at CASP14 with a median backbone error of under 1 angstrom, comparable to experimental methods, and the team released predicted structures for essentially every human protein within a year.

## Summary

AlphaFold 2 combines multiple-sequence alignments, an attention-based Evoformer network and an equivariant structure module trained end to end on the Protein Data Bank. At the blind CASP14 assessment in 2020 it achieved a median GDT of about 92 and a median backbone RMSD95 of 0.96 angstrom on the hardest targets, versus 2.8 angstrom for the next-best method.

The model provides per-residue confidence estimates (pLDDT), enabling users to distinguish reliable regions from disordered or uncertain ones. The accompanying AlphaFold Protein Structure Database, built with EMBL-EBI, released predicted structures for the human proteome and later for more than 200 million proteins.

For cancer drug discovery, AlphaFold accelerated structure-based design for targets without crystal structures and, with AlphaFold 3 (2024) extending to protein-ligand and protein-nucleic acid complexes, is now a routine part of the target-to-lead pipeline.

## Fields

- Kind: Key paper
- Last checked: 2026-09-08
- Journal: Nature
- Year: 2021
- DOI: 10.1038/s41586-021-03819-2
- Authors: Jumper J, Evans R, Pritzel A, et al.
- Findings: CASP14: median backbone RMSD95 of 0.96 angstrom (95% CI 0.85-1.16) vs 2.8 angstrom for the next-best method; Median GDT score around 92 across CASP14 targets, the first time a computational method reached experimental-grade accuracy; Per-residue confidence (pLDDT) reliably flags disordered and low-confidence regions; Predicted structures for the entire human proteome released in 2021; over 200 million proteins by 2022
- What it means: The shape of nearly every protein is now available to any researcher in seconds instead of years, which shortens the path from a cancer target to a designed molecule. It does not by itself produce drugs: binding pockets, dynamics and cellular context still need experiment.
- Caveats: Predicts single static conformations; many drug targets (kinases, GPCRs, KRAS) move between states; Accuracy is lower for proteins without evolutionary homologues, disordered regions and multi-protein complexes; Does not predict effects of point mutations or ligand binding (AlphaFold 3 and other tools partly address this); The 2021 model was released with a non-commercial licence for weights, later loosened

## Sources

- DOI: https://doi.org/10.1038/s41586-021-03819-2
- AlphaFold Protein Structure Database: https://alphafold.ebi.ac.uk

## Connected records

- technologies: [AI-driven drug & target discovery](https://onco.cc/technologies/ai-drug-design/), [AlphaFold 3](https://onco.cc/technologies/alphafold3/), [De novo designed protein binders](https://onco.cc/technologies/de-novo-protein-design/), [Structural biology infrastructure (cryo-EM, synchrotrons, AlphaFold)](https://onco.cc/technologies/structural-biology-infrastructure/)
- ideas: [Patient-level multimodal foundation models for treatment selection](https://onco.cc/ideas/idea-multimodal-foundation-model/)
- fronts: [AI & Computation](https://onco.cc/fronts/ai-computation/), [Drug Discovery Platforms](https://onco.cc/fronts/drug-discovery/)
- bottlenecks: [AI that is built but not validated or deployed](https://onco.cc/bottlenecks/b-ai-validation/), [The undruggable drivers](https://onco.cc/bottlenecks/b-undruggable-targets/), [The valley of death between lab and product](https://onco.cc/bottlenecks/b-translational-valley/)
- journals: [Nature](https://onco.cc/journals/nature/)
- people: [Faisal Mahmood](https://onco.cc/people/faisal-mahmood/)

---
JSON: https://onco.cc/api/v1/entities/paper-alphafold2-jumper-nature-2021.json