UW professor among three scientists awarded Nobel Prize in chemistry for work on proteins

A trio of scientists, including University of Washington professor David Baker, shared the Nobel Prize in chemistry on Wednesday for groundbreaking work that cracked the code of proteins, the chemical building blocks that make life possible.

The discoveries by Baker and Demis Hassabis and John M. Jumper of Google DeepMind in the United Kingdom have rapidly transformed science. Hassabis and Jumper developed a powerful computational tool that gave researchers the long-sought ability to predict how proteins twist and fold to create complex 3D structures that can block viruses, build muscle or degrade plastic.

Baker’s work opened up the ability to custom-design proteins from scratch, to create molecules with novel shapes and functions.

“We glimpsed at the beginning that it would be possible to create a whole new world of proteins that might address a lot of the problems faced by humans in the 21st century, and now it’s becoming possible,” Baker, 62, said on the phone during the announcement of the prize by the Royal Swedish Academy of Sciences in Stockholm.

Proteins are the chemical machinery of life. They make and break other molecules, they defend against pathogens, they build our bodies. They are made up of strings of amino acids, but their function is determined by the 3D form they take. Predicting those shapes from their building blocks has been one of biology’s holy grails – a 50-year quest.

Christian Anfinsen, an American scientist, won the Nobel Prize in 1972 for his discovery that the structure of a protein was determined just by its amino acid sequence.

But unraveling the nuances of how that sequence folded up into lumpy balls or intricate loops was a tough problem. In 1994, scientists began organizing a competition called the Critical Assessment of Protein Structure Prediction (CASP) – a kind of Olympics of protein folding – in which scientists would try to predict the structure of proteins whose forms had recently been decoded but not yet publicly released.

Progress was slow until 2018, when Hassabis and Jumper began to deploy tools grounded in artificial intelligence to crack the problem. The second version of their AI tool, called AlphaFold2, could predict protein structure, and it turned out it worked just as well as laborious, conventional techniques such as X-ray crystallography.

John Moult, a computational biologist at the University of Maryland and one of the founders of CASP, said it was extraordinary in mid-2020 when he began to realize that the Google DeepMind team had solved one of the biggest problems in science.

“I was there at the birth of the problem – it looked intractable, intractable, intractable. And then, suddenly, you’re there. It’s an extraordinary scientific journey,” Moult said. “You see a whole field emerging and struggling and it seems impossible, and then you get there.”

The prize is the second Nobel this year to highlight how profoundly artificial intelligence is changing society. The physics prize was awarded for contributions that undergird AI.

“This allows individual scientists to do so much more,” Hassabis, 48, said in an interview for the Nobel Prize website. “These systems, they are tools … they can’t figure out what the right question is to ask, what the right hypothesis or the right conjecture (is), and all of that has to come from a human scientist.”

Hassabis and Jumper “have designed a fantastic, ingenious neural network that solved this problem of protein structure prediction,” said Johan Åqvist, a member of the Nobel Committee for Chemistry. “This is one of the really first big scientific breakthroughs of AI.”

Hassabis, who is chief executive and co-founder of Google DeepMind, said in an interview that the next chapter of the work is already unfolding. In the next iteration, AlphaFold 3, his team models the dynamic interactions between proteins and other molecules. Eventually, he said, it may be possible to model an entire cell.

Jumper, 39, the youngest chemistry laureate in decades, said in an interview posted to the Nobel website that he had calculated that he had a 10% chance of winning the prize, and it had made for an anxious early October. He said he tried to sleep in late enough on Wednesday that he would wake up and learn whether he had won, but didn’t quite succeed.

“It’s absolutely extraordinary,” Jumper said of the honor, noting that it was particularly striking how quickly the prize was awarded, just a few years after the work was done.

Baker said he was sleeping when the committee called to inform him that he had been awarded the prize for creating a computational tool to design proteins with novel shapes and functions from scratch.

After the committee delivered the news, “my wife began screaming very loudly,” he said.

Baker also worked on the protein structure prediction problem with a tool called Rosetta. But he took it a step further, creating ways to design completely new proteins, which opens up an array of applications in vaccine design, in sensing chemicals in the environment and creating new drugs.

The ability to develop novel proteins has already proved useful. In 2017, Baker and his colleagues used Rosetta to design a protein that can detect fentanyl in the environment.

“If you could make new proteins, you could potentially solve a lot of current problems for which there aren’t proteins to deal with,” such as Alzheimer’s disease, cancer and environmental pollution, Baker said. He explained that perhaps over large stretches of time, proteins could naturally evolve to address these issues. But protein design tools allow scientists to address those challenges today.

Baker highlighted efforts to develop proteins to attack cancers while leaving other parts of the body unharmed, circumventing the downsides of current cancer treatments. A student in his lab is trying to design proteins that block the major components of snake venom, serving as an improved antivenom.

He entered college planning to study social studies and philosophy. It was not until his senior year of college that he became so intrigued by biology that he switched his major. His focus on protein folding began several years later when he joined the University of Washington as a professor.

The field has exploded, said Jon Lorsch, director of the National Institute of General Medical Sciences at the National Institutes of Health, which has funded Baker’s lab since 1995.

“We’re really just at the beginning,” Baker said. “Now, what’s tremendously exciting is to look at the vast array of problems that humans face and try and design proteins that solve these problems.”