Start date: Week 37 (9th September 2025)
Speaker 1: Patrick Bryant (Ass. Prof. at SU & DDLS Fellow)
Patrick Bryant’s research seeks to answer questions about the evolution of proteins and how this information can be used to create a new range of AI tools. Google Scholar
Title: Protein structure prediction and design
The foundation of protein design is accurate structure prediction which depends on coevolutionary patterns found in multiple sequence alignments (MSAs). Coevolution constrains the protein structure just enough for it to become predictable given a suitable MSA. However, designing new proteins poses a challenge in the absence of coevolutionary data. Therefore, protein design methodologies must incorporate learned associations between amino acid sequences and protein structures, encapsulating the rules governing the known protein space. In this lecture, you will learn more about how to design new proteins and in the practical session you will design your own peptide binder using the latest structure prediction technology.
Module 3 emphasizes structural biology covering protein structure prediction, analysis, and design. Students gain hands-on experience with AlphaFold workflows and molecular modeling integrated into their AI agents.