Project Grant K99GM164832

The National Institute of General Medical Sciences awarded the University of California, Irvine $126,489 on August 1, 2026, under the Biomedical Research and Research Training program (CFDA 93.859) to develop computational methods integrating physics-based and machine learning approaches for accelerated drug discovery.

The project addresses the high cost and extended timelines of small-molecule drug development by combining molecular dynamics-based binding free energy calculations with active learning methods to prioritize compounds for experimental testing. Aim 1 involves developing an improved active learning workflow that optimizes simultaneously for multiple objectives—potency, selectivity, and drug-likeness—while leveraging data at varying accuracy levels (docking scores and free energy calculations) to balance throughput against accuracy. Aim 2 extends the applicability of relative binding free energy calculations to larger ligand structure modifications by accounting for multiple potential binding modes. The research builds on the principal investigator's existing expertise in molecular dynamics and active learning methodologies applied to drug discovery.

Performance occurs in Irvine, California. The period of performance runs from August 1, 2026, through July 31, 2028. This is a Project Grant assistance type.