Protein Science
& Folding
Structure prediction and de novo protein design tools decode how proteins fold, function, and fail — opening new frontiers for enzyme engineering, antibody design, and disease understanding.
The Protein Folding Problem — Solved, Then Extended
For 50 years, predicting a protein's 3D structure from its amino acid sequence was one of biology's hardest unsolved problems. AI cracked it — and in doing so, opened a larger question: not just how proteins fold, but how to design entirely new ones that evolution never explored.
Our protein science programs extend state-of-the-art structure prediction into protein design, enzyme engineering, and functional modeling — asking not only "what shape does this sequence make?" but "what sequence makes the shape we need?"
Our structure models predict 3D protein conformations at sub-angstrom accuracy, faster than any experimental method — enabling structural insights across entire proteomes.
Generative protein models design novel sequences with specified functions — enzymes with new catalytic activities, antibodies with improved binding, and scaffold proteins with custom geometries.
AI models predict which proteins interact, where they bind, and how mutations disrupt those interactions — critical for understanding disease mechanisms and identifying therapeutic intervention points.