Methods and algorithms of computational mathematics and their applications

Using parallel computing in the transition interface sampling method implementation

S. V. Poluyan^a, N. M. Ershov<sup>b

a</sup> Dubna State University, Dubna, Moscow Reg.
^b Lomonosov Moscow State University, Faculty of Computational Mathematics and Cybernetics

Abstract: Numerical estimation of the reaction rate constant is an important task in the field of bioinformatics. This constant provides information on the binding kinetics of the protein complex components. One approach to perform this estimation is to apply the transition interface sampling method, which is based on the simulation of transitions between different states of the modeled system along several trajectories. Each state and transition is characterized by the estimated energy of the interaction between the components of a protein complex. In this paper, we provide a way to apply parallel computing to multiple trajectory construction. We present the principles of using MPI technology and Intel oneTBB library, along with the demonstration of various experimental results. The purpose of the study is to determine the most effective parallel programming tool for constructing multiple trajectories in the transition interface sampling method within the given computational systems.

Keywords: transition interface sampling, protein-protein interaction, reaction rate constant, MPI, Intel oneTBB.

UDC: 519.6

Received: 09.10.2024

DOI: 10.26089/NumMet.v25r434