By Karsten Eastman | Feb 12, 2026
Generative AI has already changed what is possible in protein engineering. In the last few years, the field has moved from predicting protein structures to proposing entirely new proteins, often with experimentally verified folding and binding, at a pace that was unrealistic a decade ago. Much of that progress has been driven by the ecosystem around the Baker group at the Institute for Protein Design and the broader community building on similar ideas: diffusion-based backbone generation, inverse folding, and integrated structure–sequence pipelines that can be conditioned on a target.