Researchers at Columbia University report a template-based method for prediction of protein-ligand interactions. In tests, this method compares favorably with existing template-based methods and other computational approaches. Of note is that outcomes suggest the correct identification of off-target drug binding sites and accurate predictions of ligand-protein binding in cases when proteins exhibit a weak sequence similarity to the template. This is one of eleven 2017 papers, cumulatively supported by 20 different NIH-research grants, reporting computational results enabled by the ORIP-awarded compute cluster (S10OD012351) and storage system (S10OD021764).
Strategic Plan Section
Accelerating Research Discoveries by Providing Access to State-of-the-Art Instrumentation - Continue to accelerate research discoveries by providing access to state-of-the-art instrumentation