The 2024 prize in chemistry has become another landmark moment for artificial intelligence in science. The Royal Swedish Academy of Sciences awarded half of the prize to Demis Hassabis and John M. Jumper of Google DeepMind for using AI to predict protein structures, while the other half went to David Baker of the University of Washington for computational protein design.
The winners will share a prize pot of 11 million Swedish kronor ($1 million). More than the money, the award signals how central computational biology has become to modern research, especially in work that depends on understanding how proteins take shape and behave.
Why protein structure matters
Proteins are basic to life, but knowing that a protein exists is not enough to understand what it does. Scientists also need to know its structure. That structure can determine how a protein works, what it interacts with, and how it might be used or targeted in future research.
For a long time, solving that structure was a slow and difficult problem. According to the source article, figuring out the structure of each type of protein once took months or years. That made protein research a bottleneck: the biological question might be clear, but the structural answer could take a very long time to obtain.
Computational tools change the pace of that work. When researchers can predict a protein’s structure far faster, they can ask more questions, compare more possibilities, and move more quickly toward practical uses. The potential applications named in the source include more efficient vaccines, faster research on cures for cancer, and completely new materials.
AlphaFold and the protein-folding problem
Demis Hassabis, the cofounder and CEO of Google DeepMind, and John M. Jumper, a director at the same company, were recognized for their work on artificial intelligence that predicts the structures of proteins. Their best-known system is AlphaFold.
In 2020, AlphaFold solved a problem scientists had worked on for decades: predicting the three-dimensional structure of a protein from a sequence of amino acids. That is the central challenge behind protein folding. A sequence alone is not the full story; the scientific value comes from understanding the shape that sequence forms.
AlphaFold has since been used to predict the shapes of all proteins known to science. Google DeepMind has also released the source code and database of its results to scientists for free, making the tool’s output broadly available for research.
The latest model, AlphaFold 3, extends the work beyond proteins. It can predict the structures of DNA, RNA, and molecules like ligands, which the source describes as essential to drug discovery. That broadens the system’s relevance from one central biological puzzle toward a wider set of molecular questions.
"I’ve dedicated my career to advancing AI because of its unparalleled potential to improve the lives of billions of people," said Demis Hassabis. "AlphaFold has already been used by more than two million researchers to advance critical work, from enzyme design to drug discovery. I hope we'll look back on AlphaFold as the first proof point of AI's incredible potential to accelerate scientific discovery," he added.
David Baker’s work on designing proteins
The other half of the prize went to David Baker, a professor of biochemistry at the University of Washington. His recognition centers on computational protein design, a related but distinct challenge from structure prediction.
Prediction asks what shape a given amino acid sequence will form. Design works in the other direction: researchers may have a structure or function in mind and need tools that can help identify sequences that could produce it. That shift matters because it moves computation from explaining existing biology toward helping researchers imagine new proteins.
Baker has created several AI tools for designing and predicting protein structures. The source names a family of programs called Rosetta, as well as an open-source AI tool called ProteinMPNN created by his lab in 2022.
ProteinMPNN helps researchers who already have a specific protein structure in mind find amino acid sequences that fold into that shape. The source also says the tool could help researchers discover previously unknown proteins and design entirely new ones.
In late September, Baker’s lab announced it had developed custom molecules that allow scientists to precisely target and eliminate proteins associated with diseases in living cells. That announcement fits the same broader theme: using computational design to create molecular tools that can act on specific biological targets.
“[Proteins] evolved over the course of evolution to solve the problems that organisms faced during evolution. But we face new problems today, like covid. If we could design proteins that were as good at solving new problems as the ones that evolved during evolution are at solving old problems, it would be really, really powerful,” Baker told MIT Technology Review in 2022.
What the award says about AI in science
The chemistry award is described in the source as a second Nobel win for AI. In this case, the recognition is not for a chatbot or a consumer product, but for systems that help scientists work with some of the most important structures in biology.
The common thread across AlphaFold, Rosetta, and ProteinMPNN is that they turn extremely difficult molecular questions into problems computers can help researchers explore. They do not make protein science simple. They do, however, reduce the time and friction involved in asking structural and design questions.
That is why the prize has implications beyond the individual winners. The work points to a scientific workflow in which AI tools help narrow possibilities, generate candidates, and make complex biological systems easier to investigate.
The source article frames the impact as enormous because protein structure sits so close to many areas of research and drug development. If scientists can understand proteins faster and design new ones with greater precision, the downstream possibilities include vaccines, cancer research, drug discovery, enzyme design, disease-related protein targeting, and new materials.
The Nobel Prize in chemistry therefore recognizes more than one breakthrough. It recognizes a change in how researchers can approach proteins: not only by observing and testing what nature has already produced, but also by using computational tools to predict, design, and explore molecular forms that could support future scientific work.