Application of deep learning to the protein structure prediction: the tale of a “gigantic leap”

Abstract: Knowing the three-dimensional structure of a protein is of outmost importance for understanding its function and evolutionary relationships with other members of the same protein family. However, the determination of 3D protein structures by experimental methods is not straightforward. For this reason, several computational methods have been developed in the past decades to predict the structure of proteins from their sequences. This field has experienced tremendous development in the last 4 years, when deep learning algorithms have made it possible to obtain structural predictions at a level of accuracy approaching that of experimental studies. In this seminar, the evolution of protein structure prediction methods will be shown, with special reference to the development of artificial intelligence methods, and the implications of these new approaches for knowledge not only at the level of protein structure and function, but also in terms of knowledge of the interactions with other proteins and/or macromolecules inside and outside of cells, and, prospectively, for applying this knowledge in various fields, including rational drug design.

Thursday 09/03/2023 ore 13:00 Aula A CU010 Prof. Anna Marabotti, Department of Chemistry and Biology "A. Zambelli" University of Salerno

© Università degli Studi di Roma "La Sapienza" - Piazzale Aldo Moro 5, 00185 Roma