TAPS: A Traveling-Salesman Based Automated Path Searching Method for Functional Dynamics of Complex Biomolecular Systems

Abstract

Dissecting the transitions among different metastable states of biomolecular systems is crucial for understanding their function. Nonetheless, as the transitions for complex biomolecular systems may occur at timescale much longer than the affordable length of MD simulations, various path methods have been developed for efficiency gain. Among them, path searching methods aim to locate the minimum free energy paths (MFEPs) connecting the known stable states. However, existing path searching methods typically perform local sampling around the path nodes in a pre-selected collective variable (CV) space, which limited their overall efficiency. Recently, we developed a Traveling-salesman based Automated Path Searching Method (TAPS). By using the path-collective variable (PCV), computed on the reference path, as a temporary coordinate system in each of its iteration, TAPS avoid the non-trivial a priori choice of the CVs. Meanwhile, TAPS adopts non-local perpendicular sampling to accelerate the search and uses a traveling-salesman scheme to solve the issue of node reordering brought by the perpendicular sampling. In this chapter, we describe in detail the procedure of TAPS optimization through the examples of the pentapeptide Met-enkephalin and the Mitogen-activated protein kinase kinase MEK1.