Machine Learning Directed Optimization of Classical Molecular Modeling Force Fields

Abstract

Accurate force fields are necessary for predictive molecular simulations. However, developing force fields that accurately reproduce experimental properties is challenging. Here, we present a machine learning directed, multiobjective optimization workflow for force field parameterization that evaluates millions of prospective force field parameter sets while requiring only a small fraction of them to be tested with molecular simulations. We demonstrate the generality of the approach and identify multiple low-error parameter sets for two distinct test cases: simulations of hydrofluorocarbon (HFC) vapor-liquid equilibrium (VLE) and an ammonium perchlorate (AP) crystal phase. We discuss the challenges and implications of our force field optimization workflow.

References in corpus (8)

• Scikit-learn: Machine Learning in Python
• Canonical sampling through velocity-rescaling
• Machine Learning Force Fields
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Machine learning for molecular simulation
• Simple Data and Workflow Management with the signac Framework
• Machine learning forces trained by Gaussian process in liquid states: Transferability to temperature and pressure
• EZFF: Python Library for Multi-Objective Parameterization and Uncertainty Quantification of Interatomic Forcefields for Molecular Dynamics

Cited by in corpus (4)

• Machine-learned molecular mechanics force field for the simulation of protein-ligand systems and beyond
• On the design space between molecular mechanics and machine learning force fields
• Accelerated Bayesian Inference for Molecular Simulations using Local Gaussian Process Surrogate Models
• relentless: Transparent, reproducible molecular dynamics simulations for optimization