The Year Proteins Won the Nobel — and Why Community Labels Matter

If you care about proteins, October 9, 2024 felt like a page turning. The Nobel Committee recognized a trio whose work reframed the possible: David Baker for computational protein design, and Demis Hassabis & John Jumper for AI structure prediction. It’s a signal that the center of gravity in molecular science has shifted from measuring what nature gives us to reasoning about what nature might allow.

For the Dark Protein Project, the headline is bigger than awards. It’s a mandate: if structure prediction and design are mainstream, then the dark proteome — the hundreds of millions of proteins with unclear roles — is the next frontier. Our contributors label pockets, interfaces, and catalytic patterns that make AI predictions actionable. Human judgment still distinguishes a generic cleft from a ligandable site; a plausible fold from a functional surface.

Later this month is International Open Access Week (Oct 21–27), a reminder that discoveries should be findable, usable, and reusable. Our take: the future belongs to open models + open data + open labels: a three-part engine that turns AI “maybes” into biological “likely.” We’re here for it.