Description:

Title: Computational Molecular Pharmacology.

Abstract: Computer modeling of structural, thermodynamic, and pharmacodynamic properties can accelerate drug design by inspiring new designs and prioritizing those that are most likely to be successful. My research group has recently made significant progress in binding pose, affinity, and signaling efficacy prediction. Based on a statistical mechanics framework that I derived, we have developed a method that exploits multiple rigid protein conformations to quickly compute protein-ligand binding poses and free energies. For a series of 58 ligands from a drug lead optimization campaign against the SARS-CoV-2 main protease, the method achieves a Pearson R of 0.52 and RMSE of 1.24 kcal/mol in comparison to experiment. In what could be a significant breakthrough in GPCR signaling, we have also developed a computational method that combines molecular simulation and machine learning to model structural mechanisms of activation and calculate the signaling efficacy along multiple pathways. It computes the signaling efficacy of G protein signaling along multiple subtypes and of arrestin recruitment for a diverse set of mu opioid receptor and cannabinoid receptor 2 ligands with a mean absolute error of less than 20%. I am excited about opportunities to further validate these technologies and apply them to accelerating the discovery of safe and effective therapeutics.

Speaker: David Minh, PhD; Robert E. Frey, Jr. Endowed Chair in Chemistry, Illinois Tech; Co-founder and Chief Scientific Officer, Biagon Inc.