SCORING NONCOVALENT PROTEIN-LIGAND INTERACTIONS - A CONTINUOUS DIFFERENTIABLE FUNCTION TUNED TO COMPUTE BINDING AFFINITIES

Exploitation of protein structures for potential drug leads by molecul ar docking is critically dependent on methods for scoring putative pro tein-ligand interactions. An ideal function for scoring must exhibit p redictive accuracy and high computational speed, and must be tolerant of variations in the relative protein-ligand molecular alignment and c onformation. This paper describes the development of an empirically de rived scoring function, based on the binding affinities of protein-lig and complexes coupled with their crystallographically determined struc tures. The function's primary terms involve hydrophobic and polar comp lementarity, with additional terms for entropic and solvation effects. The issue of alignment/conformation dependence was solved by construc ting a continuous differentiable nonlinear function with the requireme nt that maxima in ligand conformation/alignment space corresponded clo sely to crystallographically determined structures. The expected error in the predicted affinity based on cross-validation was 1.0 log unit, The function is sufficiently fast and accurate to serve as the object ive function of a molecular-docking search engine. The function is par ticularly well suited to the docking problem, since it has spatially n arrow maxima that are broadly accessible via gradient descent.