Why Protein Folding Matters for Medicine
Every biological process in your body ultimately comes down to proteins and how they fold. AlphaFold 2 solved this 50-year grand challenge. AlphaFold 3 extends that revolution to the entire molecular biology toolkit.
What's New in AlphaFold 3
AlphaFold 3 predicts the structure of protein-DNA, protein-RNA, and protein-small molecule complexes — the critical input for drug discovery: how does a candidate drug molecule bind to its protein target?
"We estimate AlphaFold 3 can reduce the initial hit-identification phase of drug discovery from 2-3 years to under 6 months for many target classes." — DeepMind Biology Team, 2026
Real-World Impact Already Visible
Three new antibiotics targeting drug-resistant bacteria have entered Phase 1 clinical trials in 2026, with binding predictions validated by AlphaFold 3.
Frequently Asked Questions
What is AlphaFold 3?
AlphaFold 3 is DeepMind's protein structure prediction model that extends AlphaFold 2 to predict structures of DNA, RNA, and small molecules alongside proteins, enabling comprehensive modelling of biological interactions.
How is AlphaFold used in drug discovery?
AlphaFold accelerates drug discovery by predicting 3D protein structures to identify drug targets and design molecules that bind to specific protein sites.
Is AlphaFold accurate?
AlphaFold 2 achieved near-experimental accuracy for single-chain protein structures. AlphaFold 3 extends accuracy to protein complexes, DNA-protein interactions, and ligand binding predictions.