Abstract:Molecular dynamics simulations of biomolecular processes are often discussed in terms of diffusive motion on a low-dimensional free energy landscape F(𝒙). To provide a theoretical basis for this interpretation, one may invoke the system-bath ansatz á la Zwanzig. That is, by assuming a time scale separation between the slow motion along the system coordinate x and the fast fluctuations of the bath, a memory-free Langevin equation can be derived that describes the system's motion on the free energy landscape F(𝒙), which is damped by a friction field and driven by a stochastic force that is related to the friction via the fluctuation-dissipation theorem. While the theoretical formulation of Zwanzig typically assumes a highly idealized form of the bath Hamiltonian and the system-bath coupling, one would like to extend the approach to realistic data-based biomolecular systems. Here a practical method is proposed to construct an analytically defined global model of structural dynamics. Given a molecular dynamics simulation and adequate collective coordinates, the approach employs an "empirical valence bond"-type model which is suitable to represent multidimensional free energy landscapes as well as an approximate description of the friction field. Adopting alanine dipeptide and a three-dimensional model of heptaalanine as simple examples, the resulting Langevin model is shown to reproduce the results of the underlying all-atom simulations. Because the Langevin equation can also be shown to satisfy the underlying assumptions of the theory (such as a delta-correlated Gaussian-distributed noise), the global model provides a correct, albeit empirical, realization of Zwanzig's formulation. As an application, the model can be used to investigate the dependence of the system on parameter changes and to predict the effect of site-selective mutations on the dynamics.

Ab initio generalized Langevin equation

Ab Initio Generalized Langevin Equation

Coarse-graining conformational dynamics with multi-dimensional generalized Langevin equation: how, when, and why

Internal noise-driven generalized Langevin equation from a nonlocal continuum model

Internal noise driven generalized Langevin equation from a nonlocal continuum model

Machine Learning of coarse-grained Molecular Dynamics Force Fields

Efficient Calculation of the Lattice Thermal Conductivity by Atomistic Simulations with Ab Initio Accuracy

Semi-classical generalized Langevin equation for equilibrium and nonequilibrium molecular dynamics simulation

Global Langevin model of multidimensional biomolecular dynamics

Machine Learning Exchange Fields for Ab-initio Spin Dynamics

Machine learning assisted coarse-grained molecular dynamics modeling of meso-scale interfacial fluids

Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields

ML-GLE: A machine learning enhanced Generalized Langevin equation framework for transient anomalous diffusion in polymer dynamics

In operando active learning of interatomic interaction during large-scale simulations

Modeling surface vibrations and their role in molecular adsorption: a generalized Langevin approach

Molecular Dynamics with On-the-Fly Machine Learning of Quantum-Mechanical Forces

Machine learning force-field models for metallic spin glass

Large-scale efficient Langevin dynamics, and why it works

Efficient Machine Learning Force Field for Large-Scale Molecular Simulations of Organic Systems

Machine Learning Inter-Atomic Potentials Generation Driven by Active Learning: A Case Study for Amorphous and Liquid Hafnium dioxide

DiffGLE: Differentiable Coarse-Grained Dynamics using Generalized Langevin Equation