FFLUX molecular simulations driven by atomic Gaussian process regression models

doi:10.1002/jcc.27323

FFLUX molecular simulations driven by atomic Gaussian process regression models

2024-02-12

3

PRE

AI

Y

Yulian T. Manchev

P

Paul L. A. Popelier *

DOI：10.1002/jcc.27323

Original

Original request for help

Save

Abstract

En 中文

Machine learning (ML) force fields are revolutionizing molecular dynamics (MD) simulations as they bypass the computational cost associated with ab initio methods but do not sacrifice accuracy in the process. In this work, the GPyTorch library is used to create Gaussian process regression (GPR) models that are interfaced with the next-generation ML force field FFLUX. These models predict atomic properties of different molecular configurations that appear in a progressing MD simulation. An improved kernel function is utilized to correctly capture the periodicity of the input descriptors. The first FFLUX molecular simulations of ammonia, methanol, and malondialdehyde with the updated kernel are performed. Geometry optimizations with the GPR models result in highly accurate final structures with a maximum root-mean-squared deviation of 0.064 & Aring; and sub-kJ mol(-1) total energy predictions. Additionally, the models are tested in 298 K MD simulations with FFLUX to benchmark for robustness. The resulting energy and force predictions throughout the simulation are in excellent agreement with ab initio data for ammonia and methanol but decrease in quality for malondialdehyde due to the increased system complexity. GPR model improvements are discussed, which will ensure the future scalability to larger systems.

Keywords:

FFLUX

Gaussian process regression

Interacting Quantum Atoms (IQA)

machine learning

molecular dynamics

multipole moments

QTAIM

Quantum chemical topology (QCT)

FFLUX molecular simulations driven by atomic Gaussian process regression models

Abstract

Journal

Organization