AlphaFold
DeepMind's system for predicting protein structure from sequence, which solved a 50-year-old problem in biology.
Why It Was Important
At CASP14 in 2020 AlphaFold 2 predicted protein structures to near-experimental accuracy, ending a grand challenge that had stood since the 1970s. DeepMind released predicted structures for nearly every known protein, transforming structural biology overnight, and the work brought Demis Hassabis and John Jumper the 2024 Nobel Prize in Chemistry — the clearest evidence to date that deep learning can deliver fundamental scientific results, not just commercial products. AlphaFold 3 followed in 2024, extending prediction from proteins alone to their complexes with DNA, RNA and small molecules — the form most drug discovery actually needs.
Who Invented It
John Jumper, Demis Hassabis
The DeepMind protein-structure team.
Applications
- Structural Biology
- Drug Discovery
- Enzyme Design
- Disease Research
- Protein–Ligand Complexes
Key Papers
- Highly Accurate Protein Structure Prediction with AlphaFold
John Jumper · Nature · 2021