Subscribe
Three scientists will share the 2024 Nobel Prize in Chemistry for their work on proteins, the molecules that drive chemical reactions and perform essential functions.
Demis Hassabis and John M Jumper from Google DeepMind have been awarded the prize for using artificial intelligence (AI) to predict the three-dimensional (3D) structure structure of almost all known proteins. Meanwhile, David Baker from the University of Washington has been honoured for designing a new protein that does not exist in nature.
Often referred to as the building blocks of life, proteins are made from 20 different amino acids. The sequence of these amino acids twists and folds to form the 3D structure of a protein, which dictates its function. For example, proteins can act as hormones, motors that power muscles, signalling molecules, antibodies and the building blocks of various tissues.
“One of the discoveries being recognised this year concerns the construction of spectacular proteins. The other is about fulfilling a 50-year-old dream: predicting protein structures from their amino acid sequences. Both of these discoveries open up vast possibilities,” Heiner Linke, chair of the Nobel Committee for Chemistry, said in a statement.
Over the past five decades, scientists have struggled to predict the structure of proteins based solely on their amino acid sequences. With AI, Hassabis and Jumper were able to predict the structure of almost all 200 million known proteins. Prior to this, researchers relied on X-ray crystallography to decipher the 3D structures of proteins.
The AI model was first trained on all known amino acid sequences with determined protein structures. It was then tested on amino acid sequences whose 3D structures were unknown.
The model searches databases for similar sequences and analyses which parts have remained unchanged through evolution. It also identifies which amino acids pair up, as they tend to co-evolve. Through further analysis, the tool predicts a protein’s 3D structure.
In 2020, the 50-year struggle came to an end when the AI model performed on par with X-ray crystallography. By October 2024, AlphaFold2, the AI model, had been used by more than two million people across 190 countries.
Baker, on the other hand, developed computational tools to design proteins that had not previously existed, many of which have entirely new functions. The field of protein design gained momentum towards the end of the 1990s.
The first such new protein was Top7, published in 2003. It contained 93 amino acids, making it the largest protein ever designed from scratch.
New proteins with unique functions have diverse applications, from new nanomaterials and targeted pharmaceuticals to faster vaccine development, sensitive sensors and a greener chemical industry.
