Extended diffusion theory: Recovering dynamics from biased/accelerated molecular simulations

Area of Science:

• Computational chemistry
• Molecular dynamics
• Theoretical physics

Background:

• Classical Molecular Dynamics (MD) simulations are limited to microsecond timescales, hindering the study of slow molecular phenomena.
• Biased and accelerated simulations explore conformational space rapidly but lose crucial dynamic information.
• Recovering dynamics from these simulations remains a significant challenge in computational science.

Purpose of the Study:

• To present a novel method for extracting dynamical properties from biased/accelerated simulations.
• To extend Diffusion Theory (DT) by incorporating rotational contributions and a unified mobility tensor approximation.
• To demonstrate the practical application of the extended Diffusion Theory (EDT) in recovering molecular dynamics.

Main Methods:

• The Smoluchowski equation within Diffusion Theory (DT) was employed.
• DT was extended (EDT) to include mixed rotation-translation and rotation-rotation terms in mobility tensors.
• A unified and regularized Rotne-Prager-Yamakawa approximation was utilized for mobility tensors.
• The method was applied to a peptide in explicit water using Replica Exchange Molecular Dynamics (REMD) data.

Main Results:

• The extended Diffusion Theory (EDT) successfully recovered dynamical properties (time correlation functions) from REMD simulations.
• EDT results favorably compared with those from standard MD simulations.
• The method demonstrated its ability to extract long-time dynamics from simulations that typically lose this information.

Conclusions:

• Extended Diffusion Theory (EDT) provides a robust solution for recovering long-time molecular dynamics from biased/accelerated simulations.
• EDT is a valuable tool for systems where direct MD simulations are computationally infeasible due to timescale limitations.
• This approach bridges the gap between conformational sampling and accurate dynamic property extraction.