Evaluation on the performance of MM/PBSA in nucleic acid-protein systems
Abstract The molecular mechanics/Poisson-Boltzmann surface area (MM/PBSA) method has been widely used in predicting the binding affinity among the ligands, the proteins and the nucleic acids. However, the accuracy of the predicted binding energy by the standard MM/PBSA is not always good, especially in highly charged systems. In this work, we take the protein-nucleic acid complexes as an example, and showed that the use of screening electrostatic energy (instead of coulomb electrostatic energy) in molecular mechanics can greatly improve the performance of MM/PBSA. In particular, the Pearson correlation coefficient of dataset II in the modified MM/PBSA (i.e., screening MM/PBSA) is about 0.52, much better than that (<0.33) in the standard MM/PBSA. Further, we also evaluate the effect of the solute dielectric constant and the salt concentration on the performance of the screening MM/PBSA. The present study highlights the potential power of the screening MM/PBSA for predicting the binding energy in highly charged bio-systems.