Nobel Prize winner John Jumper, SM’12, PhD’17, began his academic career as a physicist, but turned to chemistry after working on computer programs to model proteins. He pursued a PhD in theoretical chemistry at the University of Chicago, where he studied protein folding with Karl Freed, the Henry G. Gale Distinguished Service Professor Emeritusin Chemistry, and Tobin Sosnick, the William B. Graham Professor in Biochemistry and Molecular Biology.
Although modern DNA sequencing can identify the amino acid sequences that make up proteins, determining how those sequences fold into three-dimensional structures had long been slow, costly, and difficult. Jumper’s doctoral research used machine learning to model the protein-folding process, laying the groundwork for his future role at Google DeepMind.
At DeepMind, Jumper led efforts to predict protein structures directly from amino acid sequences. In 2020, he and his team unveiled AlphaFold 2, an artificial intelligence system that predicts protein structures and estimates the confidence of each prediction. Trained on a database of roughly 200,000 experimentally determined protein structures, AlphaFold 2 has transformed biological research by making high-quality structural predictions widely accessible. Although not perfect in every prediction, the system has been cited in thousands of scientific papers and has accelerated research on newly emerging viruses, drug discovery, and other areas of biology. For this breakthrough, Jumper and his DeepMind colleague Demis Hassabis each received one-quarter of the 2024 Nobel Prize in Chemistry.