Publication

Abstract : Two different scoring functions, Hirshfeldfingerprint-based scoring (HFBS) and molecular operating en-vironment (MOE), and the kernel energy meth od (KEM)along with counterpoise (CP)-corrected approach were usedto estimate the binding energies of protein–ligand complexesand tested against a series of inhibitors of human aldose re-ductase enzyme. The new scoring function, HFBS, is based onHirshfeld fingerprints, which are 2D histogram plots of thedistances from the molecular Hirshfeld surface to the nearestatomic nuclei inside versus outside the surface and are highlysensitive to the immediate environment of the molecule. TheHirshfeld surface plotted over the ligand molecule helped tovisualize the contacts with the active site residues and sol-vents, which were then taken into account for interaction en-ergy calculations. Application of KEM-assisted CP-correctedapproach facilitated an efficient way of calculating interactionenergies in protein complex systems. Interaction energies cal-culated using MP2/6-31G(d) level of theory allowed us torank the ligands by potency. We find that both the KEM-assisted CP-corr ected interaction energies and the scoringfunctio ns used here predict comparable ran kings for thestrength of binding of the series of ligands as docked to the active site of the protein, which are also in good agreementwith the experimental binding affinities in this case.