Fast and accurate estimation of protein-ligand binding remains a challenge in drug discovery. A novel, directional contact model based on a simple physical model has been developed to build an empirical scoring function that captures the essential elements of the non-bonded interactions including hydrogen bonding, electrostatic, non-polar interaction, and the desolvation effect. For interatomic interactions the directional contact model explicitly represents an atom's hybridization state and its molecular orbital geometry. The strength of interactions is determined by a simple charge model that captures the polarity of the atom, the contact distance and the direction defining the interaction geometry. This directional contact model was used to derive a surface area based implicit solvation model to capture electrostatic and entropy effects that contribute towards the desolvation effects of the protein and the ligand during the binding process. The terms of the scoring function and their weights were derived using an implementation of support vector bounding methodology by training on 203 protein-ligand complex X-ray structures spanning 10 protein families and 100 renin inhibitors with known activities. The scoring function thus derived was used to predict affinities of protein-ligand complexes from 6 protein families not in the original training set.

CONTOUR™: A new directional contact based scoring function for capturing protein-ligand binding interactions

243rd National Meeting (2012)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.