How AI Solved Biology鈥檚 Biggest Mystery
Machine learning cracked the protein-folding problem and opened up new possibilities in drug discovery.
The 2024 Nobel Prize in chemistry recognized Demis Hassabis, John Jumper and David Baker for using machine learning to tackle one of biology鈥檚 biggest challenges: predicting the 3D shape of proteins and designing them from scratch.
This award stood out because it honored research that originated at a tech company: DeepMind, an AI research startup that was acquired by Google in 2014. Most previous chemistry Nobel Prizes have gone to researchers in academia. Many laureates went on to form startup companies to further expand and commercialize their groundbreaking work鈥攆or instance, CRISPR gene-editing technology and quantum dots鈥攂ut the research, from start to end, wasn鈥檛 done in the commercial sphere.
Although the Nobel Prizes in physics and chemistry are awarded separately, there is a fascinating connection between the winning research in those fields in 2024. The physics award went to two computer scientists who laid the foundations for machine learning, while the chemistry laureates were rewarded for their use of machine learning to tackle one of biology鈥檚 biggest mysteries: how proteins fold.
The 2024 Nobel Prizes underscore both the importance of this kind of artificial intelligence and how science today often crosses traditional boundaries, blending different fields to achieve groundbreaking results.
The challenge of protein folding
Proteins are the molecular machines of life. They make up a significant portion of our bodies, including muscles, enzymes, hormones, blood, hair and cartilage.
Understanding proteins鈥� structures is essential because their shapes determine their functions. Back in 1972, Christian Anfinsen won the Nobel Prize in chemistry for showing that the sequence of a protein鈥檚 amino acid building blocks dictates the protein鈥檚 shape, which, in turn, influences its function. If a protein folds incorrectly, it may not work properly and could lead to diseases such as Alzheimer鈥檚, cystic fibrosis or diabetes.
A protein鈥檚 overall shape depends on the tiny interactions, the attractions and repulsions, between all the atoms in the amino acids it is made of. Some want to be together, some don鈥檛. The protein twists and folds itself into a final shape based on many thousands of these chemical interactions.
For decades, one of biology鈥檚 greatest challenges was predicting a protein鈥檚 shape based solely on its amino acid sequence. Although researchers can now predict the shape, we still don鈥檛 understand how the proteins maneuver into their specific shapes and minimize the repulsions of all the interatomic interactions in a few microseconds.
To understand how proteins work and to prevent misfolding, scientists needed a way to predict the way proteins fold, but solving this puzzle was no easy task.
In 2003, University of Washington biochemist David Baker wrote Rosetta, a computer program for designing proteins. With it he showed it was possible to reverse the protein-folding problem by designing a protein shape and then predicting the amino acid sequence needed to create it.
It was a phenomenal jump forward, but the shape chosen for the calculation was simple, and the calculations were complex. A major paradigm shift was required to routinely design novel proteins with desired structures.