Title: Improving drug discovery through molecular simulation
Speaker: Alex Dickson, Michigan State University
Hosted by: Pratyush Tiwary
Abstract:
Despite decades of advances in computing, genomics and structural biology, we are still unable to predict which candidates will be successful during drug development. This greatly increases the cost of bringing a new drug to market, which contributes to the rising costs of health care. The Dickson Lab at MSU works to develop new computational methodologies for modeling molecular interactions that are relevant to drug activity. This includes novel algorithms to simulate ligand unbinding events that typically occur on timescales that are millions of times longer than typical molecular dynamics simulations. From these trajectories we can identify and analyze the transition states of the ligand release pathway that determine the kinetics of ligand binding and release, which are necessary to model drug action in the body. This talk will also present a general scheme to predict molecular properties such as solubility and the partition coefficient that can be combined with (un)binding rates as inputs to holistic models of drug action in the body. Together these developments aim to improve our ability to predict successes and failures early in the drug development pipeline.