An accelerated molecular dynamics study for investigating protein pathways using the bond-boost hyperdynamics method.

Abstract

Molecular dynamics (MD) simulation is an important tool for understanding protein dynamics and the thermodynamic properties of proteins. However, due to the high computational cost of MD simulations, it is still challenging to explore a wide conformational space. To solve this problem, a variety of accelerated MD (aMD) schemes have been proposed over the past few decades. The bond-boost method (BBM) is one of such aMD schemes, which expedites escape events from energy basins by adding a bias potential based on changes in bond length. In this paper, we present a new methodology based on the BBM for accelerating the conformational transition of proteins. In our modified BBM, the bias potential is constructed using the dihedral angle and hydrogen bond, which are more suitable variables to monitor the conformational change in proteins. Additionally, we have developed an efficient algorithm compatible with the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) package. The method is validated with the conformational change of ribose binding protein and adenylate kinase by comparing the conventional and accelerated MD simulation results. Based on the aMD results, the characteristics of the proteins are investigated by monitoring the conformational transition pathways. Moreover, the free energy landscape calculated using umbrella sampling confirms all the states identified by the aMD simulation are the free energy minima, and the system makes transitions following the path indicated by the free energy landscape. Our efficient approach is expected to play a key role in investigating transition pathways in a wide range of protein simulations.