Assessing the performance of docking scoring function, FEP, MM-GBSA, and QM/MM-GBSA approaches on a series of PLK1 inhibitors

Abstract

Over-expressed polo-like kinases 1, a key regulator of cell mitosis, is associated with carcinogenesis and poor prognosis. It is very necessary to develop a reliable computational affinity prediction protocol targeting PLK1. In this study, the performance of different docking scoring function, free energy perturbation, MM-GBSA and QM/MM-GBSA were evaluated. The ranking capability of FEP is the best with r_s = 0.854. However, the r_s obtained from MM-GBSA can reach 0.767, which requires only about one-eighth of the simulation time of FEP. As for the sampling method, single long molecular dynamics (SLMD) surpass the multiple short molecular dynamics (MSMD) in ranking of the 20 congeneric compounds by about 0.1 in r_s. In addition, ligands treated by QM can significantly improve the ranking performance. As for the docking scoring functions, a force field-based scoring function is more suitable for ranking congeneric compounds.