Seminar Series in Quantitative Life Sciences and Medicine

Computational methods in drug discovery – can computers truly assist medicinal chemists?

Nicolas Moitessier (McGill University)
Tuesday February 19, 12-1pm
Montreal Neurological Institute, DeGrandpre Communications Centre

Abstract: Docking small molecules to proteins or predicting “drug-likeness” have become common practice in drug discovery teams. Over the past 15 years, we have been developing a computational drug discovery platform which has been modified to address unmet needs in medicinal chemistry. Initially, our applications of docking programs to integrin antagonists, BACE-1 inhibitors, and aminoglycosides binding to bacterial RNA revealed the limitations of available docking programs, which were essentially docking flexible ligands to rigid proteins. Over the following year, we developed our own program, FITTED, implementing algorithms for protein flexibility, displaceable water molecules, and ligand-based pharmacophore-oriented docking. Other medicinal chemistry projects motivated most of the concepts and implementation within an ever-evolving docking program. We will present the development and application of medicinal chemistry-driven implementations such as methods considering drug-zinc or iron coordination and its effect on the pKa of surrounding residues, for HDAC inhibitor design and CYP inhibition prediction, routines to identify reactive groups and form bonds with a given residue to enable the development of covalent Prolyl oligopeptidase inhibitors, methods to compute transition states while docking for studying the metabolism of POP inhibitors by cytochrome P450 enzymes (CYPs) and others.