New Research by Z. Faidon Brotzakis from Skretas Lab on Targeting AR-V7 Structural Dynamics via Deep Ensemble Docking.
Traditional in-silico drug discovery struggles with partially disordered proteins like AR-V7, due to their highly dynamic structures and numerous transient binding sites. In this study, researchers introduce a deep ensemble docking pipeline that accelerates the screening of small molecules by identifying key functional regions in the AR-V7 conformational ensemble. By reducing binding site complexity by 90× and integrating machine learning with molecular docking, they boost the discovery of multi-site binders by 17×. One compound, ChEMBL22003, reduced AR-V7’s conformational entropy and altered solvent exposure at critical regions—highlighting its potential as a phase separation modulator in prostate cancer.
DOI: 10.1021/acs.jctc.5c00171
