Mohammed AlQuraishi on Machine Learning and Structural Biology

In this episode, I speak with Prof. Mohammed AlQuraishi. Mohammed is an Assistant Professor in the Department of Systems Biology at Columbia University. Mohammed gives us his unique thoughts and perspectives on a variety of problems that lie at the intersection of machine learning and structural biology.

(00:30) -- vision of Mohammed’s lab

(02:45) -- importance of abstractions to simulate a cell

(04:29) -- conceptual advances in abstractions

(06:03) -- protein folding

(07:28) -- end-to-end differentiability

(10:12) -- representations learned by protein structure deep learning models

(14:07) -- predicting higher energy states in conformational space

(15:46) -- is structure overrated?

(18:10) -- protein localization prediction

(20:37) -- pitfalls of having a programmatic view of a cell

(23:34) -- why individuals with quantitative backgrounds may find biology interesting