Understanding Peptides and Their Interactions via Deep Learning Structure Prediction
SESSION 12: NEW FRONTIERS IN COMPUTATIONAL PEPTIDE DESIGN PART 1
Thursday, June 29, 2023, at 09:15 am - 09:40 am
Modern protein structure prediction via AlphaFold and related methods has shown a remarkable ability of machine learning methods to learn chemical and evolutionary principles from experimental biology data. Because of this understanding, these models are able to accurately model the structure of protein-peptide interactions and in some cases even identify the relative strength of different binders.
In this talk, I will discuss what ideas in machine learning enabled the high accuracy of these new models, how the community is applying them to understand peptides and their interactions, and how these models may improve in the future.
John has a Ph.D. in chemistry from the University of Chicago. Before that, he worked on molecular dynamics simulations of proteins and supercooled liquids. John leads the development of next-generation AlphaFold models. He designs algorithms, acts as a domain expert, and ensures that DeepMind’s work beneficially solves challenges in protein biology.