Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
US biochemist David Baker and Google DeepMind scientists Sir Demis Hassabis and John Jumper have won the chemistry Nobel Prize jointly for their work to unlock the biological secrets of proteins that underpin life and health.
Baker took one half of the SKr11mn ($1.06mn) award for his research on computational protein design and the DeepMind duo received the other for protein structure prediction, the Nobel Assembly in Stockholm said on Wednesday.
The prize recognises big advances in techniques to understand how proteins function and interact to make living cells work. The methods, including DeepMind’s artificial intelligence-driven AlphaFold models, have raised hopes they could be powerful tools in the development of new therapies for hard-to-treat diseases.
Baker, who is director of the Institute for Protein Design at the University of Washington, had “succeeded with the almost impossible feat of building entirely new kinds of proteins”, the Nobel organisers said. Hassabis and Jumper had “developed an AI model to solve a 50-year-old problem: predicting proteins’ complex structures”.
“Both of these discoveries open up vast possibilities,” said Heiner Linke, chair of the Nobel chemistry committee.
In a call with the Nobel committee after the announcement, Baker said he was “deeply honoured” and “stood on the shoulders of giants”, given the contribution of other researchers. “Our new AI methods are much more powerful than traditional scientific model methods. I’m really excited about all the ways in which protein design can now make the world a better place in health, medicine and . . . in technology and sustainability,” he added.
Baker has since the early 2000s harnessed computer-powered design to construct novel proteins from the 20 different building blocks, known as amino acids, of which they are mostly comprised. His teams have produced new structures for use in vaccines, nano materials and tiny sensors.
In 2022, Hassabis and Jumper’s teams used the AlphaFold AI model to build the most complete and accurate database yet of almost every known protein. Covering about 200mn proteins, the breakthrough is expected to reduce significantly the time required to make biological discoveries.
Hassabis, a co-founder and chief executive of Google DeepMind, the Silicon Valley giant’s AI research arm, described the AlphaFold innovation in March as a more efficient way “to search for the needle in a haystack”.
“That’s what really what a lot of science boils down to . . . if you can capture a problem in that way, then these types of AI systems that we’re building now can be very useful.”
The third iteration of AlphaFold unveiled by DeepMind in May extends beyond proteins to look at other biochemical networks sustaining life in our bodies’ cells. AlphaFold 3 covers the DNA and RNA genetic codes as well as ligands, molecules that bind to others and can be important markers of disease.
The Nobel committee said there had already been many applications of AlphaFold models, such as vaccine design and mining the protein database for new enzymes that could degrade plastics.
DeepMind has spun off a drug discovery arm, known as Isomorphic Labs, to build on AlphaFold’s scientific breakthroughs. Hassabis told the Financial Times this year that the goal was to use the model to cut the average discovery stage — when potential drugs are identified before clinical trials — from five years to two.
“AlphaFold has given researchers the unprecedented ability to predict what proteins “look like” in three-dimensions,” said Michael Dennis, chief scientific officer of CAS, a division of the American Chemical Society. “The impact of this technology on understanding disease mechanisms and developing drugs and new therapies is immense.”
The chemistry prize is the third of the six annual Nobels that are being announced on successive weekdays. The literature award winners will be unveiled on Thursday, followed by peace on Friday and economics on Monday.
Read the full article here
