Nobel Prize in Chemistry 2024

Nobel Prize in Chemistry 2024

Author: ChemistryViews

The Nobel Prize in Chemistry 2024 has been awarded to

  • David Baker, University of Washington, Seattle, WA, USA,

for “computational protein design”, and the other half jointly to

  • Demis Hassabis, Google DeepMind, London, UK, and
  • John M. Jumper, Google DeepMind, London, UK

for “protein structure prediction” [1].

The work of the three researchers has answered a long-standing question in biology: Can we predict how a protein folds from its amino acid sequence? Their computational methods offer highly accurate models of protein folding.

 

1 Research

1.1 Protein Structure Prediction

In proteins, amino acids are linked together to form linear peptides that fold to create a three-dimensional structure, which is key for the protein’s function. Predicting protein structures from amino acid sequences is usually notoriously difficult.

In 2020, Demis Hassabis and John Jumper presented an artificial intelligence (AI) model called AlphaFold2 [2]. With its help, they have been able to predict the structure of almost all of the 200 million known proteins. Since their breakthrough, AlphaFold2 has been used by more than two million people from 190 countries.

 

1.1.1 From Chess to Proteins

Demis Hassabis, who started playing chess at the age of four and achieved master level as a 13-year-old, started a career as a programmer and game developer in his teen years. He then began exploring artificial intelligence and neuroscience, with the latter allowing him to develop better neural networks for AI. In 2010, he co-founded DeepMind, a company that developed AI models for popular board games and was sold to Google in 2014. DeepMind achieved a breakthrough in AI by beating the champion player of the board game Go [3].

Hassisbis then tackled the problem of protein folding, developing the AI model AlphaFold [4]. The model achieved nearly 60 % accuracy in a prominent protein structure prediction competition (“Critical Assessment of Protein Structure Prediction”, or CASP), which was sufficient to win, but there was still progress to be made.

 

1.1.2 Jumping in with Improvements

John Jumper first studied physics and mathematics, and he realized that knowledge of physics could help solve medical problems while working with supercomputers to simulate proteins and their dynamics. In 2011, he began his doctorate in theoretical physics.

To save computer capacity, he started developing simpler and more ingenious methods for simulating protein dynamics. In 2017, after completing his doctorate, he heard rumors that DeepMind had started to predict protein structures. He had ideas about how to improve AlphaFold and joined the team. Jumper and Hassabis co-led the work that fundamentally reformed the AI model.

The new version of the model, called AlphaFold2, was influenced by Jumper’s knowledge of proteins. The team also started to use the innovation behind the recent enormous breakthrough in AI: neural networks called transformers.

These models can find patterns in enormous amounts of data in a more flexible manner than previously possible, and efficiently determine where the focus should be to achieve a particular goal. The team trained AlphaFold2 on the vast information contained in databases of all known protein structures and amino acid sequences, and the new AI architecture started delivering good results.

In 2020, when the organizers of the CASP competition evaluated the results, they understood that the arguably biggest problem in biochemistry was solved. In most cases, AlphaFold2 performed almost as well as X-ray crystallography at determining protein structures!

 

1.2 Computational Protein Design

Proteins are generally built from the 20 natural amino acids, which can be combined to form an endless amount of different peptides and proteins. In 2003, David Baker used computational methods to design a new protein that was unlike any previously known one. Since then, his work has produced many new protein creations, including proteins that can be used in pharmaceuticals, vaccines, nanomaterials, and sensors.

 

1.2.1 Rosetta

David Baker first studied philosophy and social science. However, he took a course in evolutionary biology which led him to change directions. He began to explore cell biology and became fascinated by protein structures. He worked on protein folding, and, at the end of the 1990s, he began to develop software that could predict protein structures: Rosetta [5]. The software was used in the CASP competition in 1998 and performed well.

The success of Rosetta led to a new idea: using the software in reverse. Instead of entering amino acid sequences in Rosetta and getting protein structures out, researchers might be able to enter a desired protein structure and obtain suggestions for its amino acid sequence, which would allow them to create entirely new proteins.

 

1.2.2 New Proteins Created From Scratch

Protein design allows researchers to create tailored proteins with new or improved functions. Often, scientists “tweak” existing proteins, so they can, for example, serve as tools in the chemical industry or break down hazardous substances. However, the range of natural proteins is limited, and creating proteins “from scratch” (de novo design) can improve the potential for obtaining ones with entirely new functions.

Baker designed a protein with an entirely new structure and then had Rosetta compute which type of amino acid sequence could result in the desired protein. To do this, Rosetta searched a database of all known protein structures and looked for short fragments of proteins that had similarities with the desired structure. Using a fundamental knowledge of proteins’ energy landscape, Rosetta then optimized these fragments and proposed an amino acid sequence.

Baker introduced the gene for the proposed amino acid sequence in bacteria that produced the desired protein, and then he determined the protein structure using X-ray crystallography. It turned out that Rosetta really could construct proteins. The protein that Baker had developed, Top7, had almost exactly the desired structure, which did not exist in nature [6].

 

Laureates

David Baker, born in Seattle, Washington, USA, on October 6, 1962, studied at Harvard University, Cambridge, MA, USA, and received his Ph.D. in biochemistry in 1989 from the University of California, Berkeley, USA, in the laboratory of Randy Schekman, where he focused on protein transport and trafficking in yeast. After postdoctoral work in biophysics with David Agard at the University of California, San Francisco, USA, he joined the faculty of the Department of Biochemistry at the University of Washington School of Medicine in 1993 and became a Howard Hughes Medical Institute Investigator in 2000. Currently, Baker is Head of the Institute for Protein Design and the Henrietta and Aubrey Davis Endowed Professor of Biochemistry at the University of Washington.

Among many other honors, David Baker received the Feynman Prize in Nanotechnology in 2004, the Sackler International Prize in Biophysics in 2008, the Breakthrough Prize in Life Sciences in 2021, and the Wiley Prize in Biomedical Sciences in 2023. He was elected a Fellow of the American Academy of Arts and Sciences in 2009.

 

Demis Hassabis, born in London, UK, on July 27, 1976, studied computer science at the University of Cambridge, UK. He worked as a programmer and game designer before receiving his Ph.D. in cognitive neuroscience from University College London (UCL), UK, in 2009. Hassabis showed remarkable talent for chess from the age of four, captained many England junior teams.

Demis Hassabis is the CEO and co-founder of DeepMind, a machine learning AI startup that was founded in London in 2010 together with Shane Legg and Mustafa Suleyman. In 2014, Google purchased DeepMind. DeepMind’s tool AlphaFold aims to accurately predict protein structures.

Demis Hassabis is a Fellow of the Royal Society and has won numerous awards for his work on AlphaFold, including the Breakthrough Prize, the Wiley Prize in Biomedical Sciences, the Canada Gairdner International Award, and the Lasker Award, all in 2023. In 2017, he was included in the Time 100 list of the world’s most influential people. He was knighted for services to AI in 2024.

 

John M. Jumper, born in the USA, studied at the University of Chicago, USA, where he received his Ph.D. in 2017 for research on the use of machine learning to simulate protein folding and dynamics, co-supervised by Tobin R. Sosnick and Karl Freed. He also studied physics at the University of Cambridge, UK, where he was a Marshall Scholar, and physics and mathematics at the Vanderbilt University, Nashville, TN, USA. He is currently a senior research scientist at DeepMind Technologies, London, UK.

John Jumper has won numerous awards, including the Wiley Prize in Biomedical Sciences in 2023. The journal Nature named Jumper one of the ten “people who matter” in science in its annual list of Nature’s 10 in 2021.

 

References

[1] Website of the Nobel Foundation nobelprize.org

[2] J. Jumper, R. Evans, A. Pritzel, T. Green, M. Figurnov, O. Ronneberger, K. Tunyasuvunakool, R. Bates, A. Zidek, A. Potapenko, A. Bridgland, C. Meyer, S. A. A. Kohl, A. J. Ballard, A. Cowie, B. Romera-Paredes, S. Nikolov, R. Jain, J. Adler, T. Back, S. Petersen, D. Reiman, E. Clancy, M. Zielinski, M. Steinegger, M. Pacholska,T. Berghammer, S. Bodenstein, D. Silver, O. Vinyals, A. W. Senior, K. Kavukcuoglu, P. Kohli, D. Hassabis, Highly accurate protein structure prediction with AlphaFold, Nature 2021, 596, 583-589. https://doi.org/10.1038/s41586-021-03828-1

[3] S. Silver, T. Hubert, J. Schrittwieser, I. Antonoglou, M. Lai, A. Guez, M. Lanctot, L. Sifre, D. Kumaran, T. Graepel, T. Lillicrap, K. Simonyan, D. Hassabis, A general reinforcement learning algorithm that masters chess, shogi and Go through self-play, Science 2018, 262, 1140-1144. https://doi.org/10.1126/science.aar6404

[4] A. W. Senior, R. Evans, J. Jumper, J. Kirkpatrick, L. Sifre, T. Green, C. Qin, A. Žídek, A. W. R. Nelson, A. Bridgland, H. Penedones, S. Petersen, K. Simonyan, S. Crossan, P. Kohli, D. T. Jones, D. Silver, K. Kavukcuoglu, D. Hassabis, Improved protein structure prediction using potential from deep learning, Nature 2020, 577, 706-710. https://doi.org/10.1038/s41586-019-1923-7

[5] K. T. Simons, R. Bonneau, I. Ruczinski, D. Baker, Ab initio protein structure prediction of CASPIII targets using ROSETTA, Proteins: Struct. Funct. Genet. Suppl. 1999, 3, 171-176. https://doi.org/10.1002/(SICI)1097-0134(1999)37:3+<171::AID-PROT21>3.0.CO;2-Z

[6] B. Kuhlman, G. Dantas, G. C. Ireton, G. Varani, B. L. Stoddard, D. Baker, Design of a novel globular protein fold with atomic-level accuracy, Science 2003, 302, 1364-1368. https://doi.org/10.1126/science.108942

 

Selected Publications

Selected Publications by David Baker


Selected Publications by Demis Hassabis


Selected Publications by John Jumper


Selected Publications by Demis Hassabis and John Jumper

 

Also of Interest

 

 

Leave a Reply

Kindly review our community guidelines before leaving a comment.

Your email address will not be published. Required fields are marked *