Simulating Replica Exchange: Markov State Models, Proposal Schemes, and the Infinite Swapping Limit

Area of Science:

• Computational Chemistry
• Biophysics
• Molecular Dynamics Simulations

Background:

• Replica exchange molecular dynamics (REMD) enhances sampling of complex biomolecular systems.
• Efficient REMD implementation is crucial but challenging due to convergence issues.
• Understanding REMD cycle construction is key to improving simulation efficiency.

Purpose of the Study:

• To develop and utilize a Markov state model of replica exchange (MSMRE) for analyzing REMD efficiency.
• To investigate the impact of replica exchange cycle parameters on simulation sampling.
• To provide methods for estimating the infinite swapping limit in REMD.

Main Methods:

• Construction of a Markov state model of replica exchange (MSMRE).
• Analysis of long molecular dynamics simulations of a host-guest binding system.
• Evaluation of replica exchange attempt frequency and MD steps per cycle.
• Estimation of the infinite swapping limit using transition matrices.

Main Results:

• MSMRE effectively models replica exchange dynamics.
• The number of replica exchange attempts and MD steps per cycle significantly affect sampling efficiency.
• The study provides methods to estimate the infinite swapping limit from MSMRE and short REMD runs.

Conclusions:

• Optimizing REMD cycle parameters is essential for efficient biomolecular simulations.
• MSMRE offers a robust framework for analyzing and improving REMD strategies.
• Accurate estimation of the infinite swapping limit enhances the reliability of free energy calculations.