David Baker is a biochemist at the University of Washington and director of its Institute for Protein Design. Over more than two decades his lab has been one of the central forces in computational protein science, first for predicting the shapes of natural proteins and later for designing entirely new ones that do not exist in nature.
His group developed Rosetta, a long-running software suite for modeling and designing proteins, and in 2003 created Top7, a protein with a three-dimensional fold never before seen in living things. In 2021 the lab released RoseTTAFold, a deep-learning structure predictor that matched much of AlphaFold 2’s accuracy and was shared openly, helping spread the new methods across academia and industry. The institute went on to build generative tools such as RFdiffusion for designing custom proteins on demand.
In October 2024 Baker was awarded one half of the Nobel Prize in Chemistry for computational protein design; the other half went jointly to Demis Hassabis and John Jumper of Google DeepMind for protein structure prediction. According to his institute, his proteins have been engineered to neutralize viruses, target cancer cells, and act as catalysts, and his work contributed to a computationally designed component of a COVID-19 vaccine.
For a general reader, Baker represents the design side of the AI-and-biology story: not just predicting how natural proteins fold, but using machine learning to invent new molecules for medicine and industry.