Gregory Ross

Improving Affinity Calculations for Drug-Protein Interactions

Research Interests

A physicist by training, I am interested in a quantitive understanding of biological systems. Biology is the ultimate complex system and often the 'kitchen sink' approach is often the best way to understand a problem. I enjoy employing a wide range of techniques in my research, including statistical mechanics, information theory, machine learning and molecular simulations.

Research

Virtual screening is a vital cog in the drug discovery process. It is a computational tool used to rapidly assess which compounds from a chemical library are most likely to bind strongly to a pharmacological protein target. The chemical libraries can be huge, so each potential ligand has to be processed as fast as possible.

My research is in developing methodologies that can calculate the free energy of ligand binding accurately and rapidly. Previous research indicated that consideration of explicit water molecules in binding sites can lead to improved affinity calculations. It can also be important to consider waters when trying to design and analyse new chemical compounds to treat diseases. Because of the often dubious nature of water molecules in protein crystal structures, I developed a methodogy to accurately predict their location that was fast enough to apply to very large data sets. It's called WaterDock and more details can be found here.

Publications

G.A. Ross, G. M. Morris, P.C. Biggin, Rapid and Accurate Prediction and Scoring of Water Molecules in Protein Binding Sites, PLoS One (2012) 7. e32036.