Evaluating the Reliability of MM-PB/GB-SA Method for the Protein-Ligand Binding Free Energies Using Penicillopepsin-Inhibitor ligands

Abstract

An accurate prediction of the ligand-receptor binding free energies (ΔG) is a critical step in the early stages of rational drug design. The Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) method is a popular   approach to estimate ΔG. However, correlations between the predicted and the experimental ΔG are variable. The goal of this study is to investigate various approaches to optimize accuracy of the MM-GBSA method. A molecular dynamic (MD) simulations protocol was applied using penicillopepsin receptor against its inhibitor ligands, repeated 50 times for each complex system. After that, ΔG of the five inhibitors were predicted using MM-GBSA method. Moreover, a diverse ΔG values were calculated from the replicate MD simulations of each system. The results were showed correlations not only between the predicted and the experimental binding affinities of the systems but also between the predicted values and root-mean-square deviation. In addition, statistical analysis was evaluated the sample size.