Modern day artificial intelligence (AI) has been successful in many domains of life in making rapid sense of huge amounts of data. It is tempting to ask if AI could also be used to understand and predict the dynamics of complex molecules with millions of atoms. In this talk I will show that certain flavors of AI can indeed help us understand generic molecular dynamics (MD) and also predict it even in situations with arbitrary long memories. However this requires close integration of AI with old and new ideas in statistical mechanics. Here I will talk about such recent methods developed my group. I will demonstrate the methods on the study of ligand dissociation from flexible receptors, including benzene-T4 lysozyme complex, different ligands with the PreQ<sub>1</sub> riboswitch and Gleevec (Imatinib) from Abl kinase. I will show how these AI-augmented MD methods can predict dissociation timescales, mechanisms and resistance mutations with high accuracy and minimum human intervention in the calculation process.

Predicting drug dissociation kinetics, pathways and resistance mutations with AI-augmented molecular dynamics

ACS Fall 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.