PROJECT SUMMARY/ ABSTRACT Our ability to rationally develop new drugs or engineer proteins is limited by an inadequate understanding of conformational dynamics of biomolecules. We have previously established that proteins sample a restricted, yet functionally relevant, conformational ensemble in a crystalline environment. Unfortunately, the ensembles present themselves as unresolved, spatiotemporally averaged data in X-ray diffraction experiments. Our work has centered on overcoming this limitation by resolving ensembles with multi-temperature X-ray crystallography and multi-conformer models Our central hypothesis is that today's structural models disproportionately target the ensemble average, masking important functional and allosteric molecular mechanisms. The objective of this research program is to use multi-temperature X-ray crystallography and multi-conformer models to access multi-scale heterogeneity and shifting equilibria of proteins and protein- ligand systems. We will pursue the following specific aims: 1. Map the conformational landscapes of PTP1B and AR by Multitemperature Multiconformer X-ray crystallography (MMX). To examine functionally important loop regions, we will create and validate new automated procedures for our qFit algorithm to accurately model large backbone heterogeneity. We will apply MMX to PTP1B, a validated diabetes target, and Androgen Receptor (AR), an important prostate cancer target with new emerging drug resistance. 2. Exploit MMX models for allosteric ligand discovery in PTP1B and Androgen Receptor. We have already identified novel allosteric modulators for the diabetes therapeutic target protein tyrosine phosphatase 1B (PTP1B) by multitemperature crystallography and ligand tethering. We will determine structures and test novel ligands of AR resistance mutants. 3. Map conformational coupling at protein-ligand binding interfaces by MMX. We will create and validate new procedures for our qFit algorithm to model ligand conformational distributions and solvated amino acids, which is important for ligand optimization. We will perform large-scale validation tests of our algorithms. Our research program will have a positive impact by enabling discovery-driven, translational ligand optimization. Robust experimental and computational methods to access conformational ensembles, in a manner complimentary to NMR and molecular dynamics, that would open up a new and exciting and avenue to reveal molecular mechanisms not only from conventional synchrotron data, but also from time-resolved experiments at new XFEL lightsources and from cryoEM.