EdSr: A Novel End-to-End Approach for State-Space Sampling in Molecular Dynamics Simulation

Area of Science:

• Computational Chemistry
• Molecular Modeling
• Biophysics

Background:

• Molecular dynamics (MD) simulations are crucial for studying complex systems.
• Current MD methods are limited by small integration time steps, restricting accessible timescales.
• Efficient simulation of large biological molecules and materials requires longer time steps.

Purpose of the Study:

• To introduce a novel method, exploratory dynamics sampling with recursion (EdSr), for flexible time step adjustment in MD simulations.
• To demonstrate the capability of EdSr to overcome the time scale limitations of conventional MD integrators.
• To enhance simulation stability for long-timescale studies.

Main Methods:

• Developed EdSr based on ordinary differential equations and Taylor expansion.
• Implemented EdSr in five experimental setups: simple functions, physical models, all-atom, coarse-grained, and dissipative particle dynamics.
• Introduced rEdSr to improve stability and mitigate velocity drift in long-timescale simulations.

Main Results:

• EdSr dynamically adjusts time steps based on simulation requirements.
• EdSr successfully operated with larger time steps compared to the velocity-Verlet integrator across various systems.
• rEdSr demonstrated enhanced stability for extended simulations.

Conclusions:

• EdSr offers a promising approach for flexible time step selection in MD simulations.
• The method has the potential to significantly extend the accessible time scales in molecular simulations.
• Further development is warranted to optimize performance across all simulation systems.