David Baker

Henrietta and Aubrey Davis Endowed Professor of Biochemistry, University of Washington; Director, Institute for Protein Design. Nobel Prize in Chemistry, 2024 — for computational protein design. Panelist at Session XII.

Protein Design Platform

Uses diffusion-based generative models (conceptually similar to DALL-E for images) to generate proteins with near-perfect complementarity to virtually any target.

Key Capabilities Demonstrated

• De novo binders: TNF receptor binder at 6-picomolar affinity; cross-reacts across human, mouse, cynomolgus; survives gut proteases
• Intrinsically disordered proteins: co-diffuse target and binder → tau binders that suppress/disassemble fibrils; fused to FBOX → bioPROTAC achieving complete tau clearance
• Designed proteases: entirely new folds; enhance amide bond hydrolysis by >10 orders of magnitude; TDP-43 targeting protease
• Cancer: mesothelin binders → CAR-T; bioPROTACs targeting DNMT3A reduce T-cell exhaustion; conditionally active interferon beta (masked until tumor microenvironment)
• Neoantigen discrimination: mini-proteins that distinguish individual peptide-MHC neoantigens where conventional antibodies fail

Immunogenicity Advantage

Even if antibodies form against de novo proteins, they cannot cross-react with any human protein — eliminating autoimmunity risk. A counterintuitive safety advantage over engineered native proteins.

On Clinical Data

“If we had full clinical trial results on everything being tested all over the world, maybe we would start generating enough data… the structural data is enormous — and the clinical data is tiny, tiny, tiny.”

The Rejection Letter

Applied for a Markey Foundation fellowship and didn’t get it. A decade later they asked how not receiving the award had “negatively impacted” his career. That letter hangs in his lab.