Protein structure prediction is working out the 3D shape a protein folds into from its sequence of amino acids, a problem AI models such as AlphaFold largely cracked.
In plain words
A protein's shape decides what it does, but measuring shapes in the lab can take months or years. For over 50 years scientists tried to predict shape from sequence alone. AlphaFold 2 matched lab accuracy in most test cases in 2020, and two of its creators, Demis Hassabis and John Jumper, shared half of the 2024 Nobel Prize in Chemistry. Its database now holds over 200 million predicted structures.
A quick example
Imagine a lab studying a little-known bacterial protein. Instead of waiting months for a lab structure, it looks the protein up in the AlphaFold database and gets a predicted 3D shape to guide which experiments to run first.
Where you'll meet it
- Biology and medical research.
- Drug and enzyme design.
Related
Sources
- PaperHighly accurate protein structure prediction with AlphaFold, Jumper et al. (Nature) · read 29 Sept 2026
- DocsAlphaFold, Google DeepMind · read 29 Sept 2026
- DocsThe Nobel Prize in Chemistry 2024 press release, Nobel Prize Outreach · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.