DeepMind Researchers and David Baker Awarded 2024 Nobel Prize in Chemistry

Google DeepMind’s Demis Hassabis and John Jumper have shared the 2024 Nobel Prize in Chemistry with American biochemist David Baker.

The Royal Swedish Academy of Sciences awarded the 11 million Swedish kronor (£810,000) prize in Stockholm. Professor Baker, based at the University of Washington, received half of the award for computational protein design. Sir Demis, chief executive of Google DeepMind, and senior research scientist Dr John Jumper shared the remaining half for protein structure prediction.

Proteins govern the chemical reactions that sustain life, operating as cellular building blocks, hormones, antibodies, and molecular machines. A protein’s biological function depends entirely on its three-dimensional shape, which forms when a linear chain of amino acids folds. Determining how sequence permutations fold into physical structures occupied molecular biologists for half a century. Heiner Linke, chair of the Nobel Committee for Chemistry, noted that predicting structural configurations from sequences had been considered impossible for four decades.

DeepMind addressed the prediction challenge in 2020 with the release of AlphaFold2. The model has predicted the three-dimensional structures of virtually all 200 million catalogued proteins identified by researchers. More than two million scientists across 190 countries currently use the system. Documented applications include identifying mechanisms behind antibiotic resistance, designing targeted therapeutics, and engineering enzymes capable of breaking down plastics.

While DeepMind focused on mapping existing biological structures, Baker developed computational methods to build entirely new proteins. In 2003, Baker succeeded in assembling a novel protein structure from amino acid units that did not exist in nature. His research group has since generated custom proteins deployed in pharmaceuticals, vaccines, miniature sensors, and nanomaterials.

Together, the two methodologies establish a pipeline spanning structural mapping and computational molecular engineering. AlphaFold has charted the structural universe of known proteins. Can computational chemists now leverage these predictive foundations to construct functional, synthetic molecular machinery from scratch?