Using arti昀椀cial intelligence to create Further information three-dimensional structures AlphaFold Nobel Prize in Chemistry for the development https://alphafold.ebi.ac.uk Nobel Prize of AlphaFold https://www.nobelprize.org Publications in Nature This year’s Nobel Prize in Chemistry went in part to Demis Hass- https://www.nature.com/ articles/d41586-019-01357-6 abis and John Jumper from Google’s DeepMind research center. https://www.nature.com/arti- The two scientists were instrumental in developing AlphaFold, cles/d41586-024-03708-4 an arti昀椀cial intelligence (AI) that allows researchers worldwide to predict the three-dimensional structure of proteins based on the sequence of amino acids. This AI is also being used by several teams in the SNI network. To 昀椀nd out more about AlphaFold and its applications, we spoke to Dr. Janani Durairaj from the group led by Professor Torsten Schwede (Biozentrum) and Professor Timm Maier (Biozentrum). SNI INSight: Why is it important to ana- lyze the three-dimensional structure of a protein? Timm Maier: The three-dimensional struc- ture of a protein determines how it interacts with other molecules in its environment. This structure is a key element in the speci- 昀椀city, activity and regulation of proteins. In the case of an enzyme, for example, the shape of the binding site must perfectly match the substrate to which it binds. If the shape of an enzyme changes, the catalyzed re- action can be blocked. The three-dimensional structure is also crucial for the formation of larger protein complexes and for proteins that ful昀椀ll other functions. If it changes due Timm Maier explains why the three-dimensional struc- to mutations, chemical modi昀椀cations or en- ture of a protein is so important. vironmental in昀氀uences, this can have serious consequences for the function of the protein and thus for the entire organism. Knowledge of the protein structure SNI INSight: What did the Nobel Prize win- helps researchers when it comes to devel- ners Demis Hassabis and John Jumper oping drugs that interact with proteins and develop? either block or intensify their activity or to developing proteins with new or improved Janani Durairaj: The two researchers have functions. contributed signi昀椀cantly to the development of the arti昀椀cial intelligence AlphaFold2, which can predict the spatial structure of proteins of which only the amino acid se- quence is known. 4 SNI INSight December 2024