Improving force field accuracy by training against condensed phase mixture properties

Boothroyd, Simon; Madin, Owen; Mobley, David L.; Wang, Lee‐Ping; Chodera, John D.; Shirts, Michael R.

doi:10.26434/chemrxiv-2021-gsgr4-v3

Cited by 4 publications

(16 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Properties of pure organic liquids are commonly used to validate classical FF, serving in many cases also as parametrization targets (e.g., [75], [138], [139]). To further explore the performance of the IPA d+c model beyond vacuum and the crystalline phase, applications to the liquid phase are shown in the following.…”

Section: Pure Liquid Propertiesmentioning

confidence: 99%

“…To further explore the performance of the IPA d+c model beyond vacuum and the crystalline phase, applications to the liquid phase are shown in the following. Three benchmarks covering a wide range of systems and properties are considered for this: (i) 13 sulfur compounds taken from the publication of the OPLS4 release [140], (ii) 29 molecules containing H, C, O taken from a recent investigation of condensed-phase parametrization of OpenFF [139], and (iii) 57 organic compounds from the GROMOS 2016H66 validation [138]. Unlike the referenced FF, parametrization of the IPA d+c model did not include experimental liquid properties such as the density or the heat of vaporization.…”

Section: Pure Liquid Propertiesmentioning

confidence: 99%

“…Test systems from OpenFF. To gain a better understanding for the role of the bonded terms taken from OpenFF, results for 29 pure liquids from a recent benchmark of OpenFF are presented here [139]. The compounds contain only H, C, and O.…”

Section: Rmse For Pure Liquid Properties Of Sulfur Compoundsmentioning

confidence: 99%

“…As can be seen in Table 6, similar errors are observed for the IPA d+c model and the standard OpenFF 1.0. Re-optimization of the FF with respect to the pure liquid properties of the training set (also compounds containing only H, C, and O) improved the accuracy of OpenFF 1.0 considerably [139] (Table 6). The authors observed thereby opposing gradient components for the simultaneous optimization with respect to densities and heats of vaporization.…”

Section: Rmse For Pure Liquid Properties Of Sulfur Compoundsmentioning

confidence: 99%

“…No further liquid properties were considered in Ref. [139] (such as dielectric permittivity, thermal expansion coefficient, etc.). It would thus be interesting to see the performance of the re-optimized OpenFF (termed 'Pure only') on other properties.…”

Section: Rmse For Pure Liquid Properties Of Sulfur Compoundsmentioning

confidence: 99%

See 4 more Smart Citations

Regularized by Physics: Graph Neural Network Parametrized Potentials for the Description of Intermolecular Interactions

Thürlemann¹,

Böselt²,

Riniker³

2022

Preprint

View full text Add to dashboard Cite

Simulations with an explicit description of intermolecular forces using electronic structure methods are still not feasible for many systems of interest. As a result, empirical methods such as force fields (FF) have become an established tool for the simulation of large and complex molecular systems. However, the parametrization of FF is time consuming and has traditionally been based largely on experimental data, which is scarce for many functional groups. Recent years have therefore seen increasing efforts to automatize FF parametrization and a move towards FF fitted against quantum-mechanical reference data. Here, we propose an alternative strategy to parametrize intermolecular interactions, which makes use of machine learning and gradient-descent based optimization while retaining a functional form founded in physics. This strategy can be viewed as generalization of existing FF parametrization methods. In the proposed approach, graph neural networks are used in conjunction with automatic differentiation to parametrize physically motivated models to potential-energy surfaces, enabling full automatization and broad applicability in chemical space. As a result, highly accurate FF models are obtained which retain the computational efficiency, interpretability and robustness of classical FF. To showcase the potential of the proposed method, both a fixed-charge model and a polarizable model are parametrized for intermolecular interactions and applied to a wide range of systems including dimer dissociation curves and condensed-phase systems.

show abstract

Section: Pure Liquid Propertiesmentioning

confidence: 99%

Section: Pure Liquid Propertiesmentioning

confidence: 99%

Section: Rmse For Pure Liquid Properties Of Sulfur Compoundsmentioning

confidence: 99%

Section: Rmse For Pure Liquid Properties Of Sulfur Compoundsmentioning

confidence: 99%

Section: Rmse For Pure Liquid Properties Of Sulfur Compoundsmentioning

confidence: 99%

See 3 more Smart Citations

Regularized by Physics: Graph Neural Network Parametrized Potentials for the Description of Intermolecular Interactions

Thürlemann¹,

Böselt²,

Riniker³

2022

Preprint

View full text Add to dashboard Cite

show abstract

Open Force Field Evaluator: An Automated, Efficient, and Scalable Framework for the Estimation of Physical Properties from Molecular Simulation

Boothroyd

Wang

Mobley

et al. 2022

J. Chem. Theory Comput.

Self Cite

View full text Add to dashboard Cite

Developing accurate classical force field representations of molecules is key to realizing the full potential of molecular simulations, both as a powerful route to gaining fundamental insights into a broad spectrum of chemical and biological phenomena and for predicting physicochemical and mechanical properties of substances. The Open Force Field Consortium is an industry-funded open science effort to this end, developing open-source tools for rapidly generating new high-quality small-molecule force fields. An integral aspect of this is the parameterization and assessment of force fields against high-quality, condensed-phase physical property data, curated from open data sources such as the NIST ThermoML Archive, alongside quantum chemical data. The quantity of such experimental data in open data archives alone would require an onerous amount of human and computational resources to both curate and estimate manually, especially when estimations must be obtained for numerous sets of force field parameters. Here, we present an entirely automated, highly scalable framework for evaluating physical properties and their gradients in terms of force field parameters. It is written as a modular and extensible Python framework, which employs an intelligent multiscale estimation approach that allows for the automated estimation of properties from simulation and cached simulation data, and a pluggable API for estimation of new properties. In this study, we demonstrate the utility of the framework by benchmarking the OpenFF 1.0.0 small-molecule force field and GAFF 1.8 and GAFF 2.1 force fields against a test set of binary density and enthalpy of mixing measurements curated using the framework utilities. Further, we demonstrate the framework's utility as part of force field optimization by using it alongside ForceBalance, a framework for systematic force field optimization, to retrain a set of nonbonded van der Waals parameters against a training set of density and enthalpy of vaporization measurements.

show abstract

Regularized by Physics: Graph Neural Network Parametrized Potentials for the Description of Intermolecular Interactions

Thürlemann

Böselt

Riniker

2023

J. Chem. Theory Comput.

View full text Add to dashboard Cite

Simulations of molecular systems using electronic structure methods are still not feasible for many systems of biological importance. As a result, empirical methods such as force fields (FF) have become an established tool for the simulation of large and complex molecular systems. The parametrization of FF is, however, time-consuming and has traditionally been based on experimental data. Recent years have therefore seen increasing efforts to automatize FF parametrization or to replace FF with machine-learning (ML) based potentials. Here, we propose an alternative strategy to parametrize FF, which makes use of ML and gradient-descent based optimization while retaining a functional form founded in physics. Using a predefined functional form is shown to enable interpretability, robustness, and efficient simulations of large systems over long time scales. To demonstrate the strength of the proposed method, a fixed-charge and a polarizable model are trained on ab initio potential-energy surfaces. Given only information about the constituting elements, the molecular topology, and reference potential energies, the models successfully learn to assign atom types and corresponding FF parameters from scratch. The resulting models and parameters are validated on a wide range of experimentally and computationally derived properties of systems including dimers, pure liquids, and molecular crystals.

show abstract

Improving force field accuracy by training against condensed phase mixture properties

Cited by 4 publications

References 68 publications

Regularized by Physics: Graph Neural Network Parametrized Potentials for the Description of Intermolecular Interactions

Regularized by Physics: Graph Neural Network Parametrized Potentials for the Description of Intermolecular Interactions

Open Force Field Evaluator: An Automated, Efficient, and Scalable Framework for the Estimation of Physical Properties from Molecular Simulation

Regularized by Physics: Graph Neural Network Parametrized Potentials for the Description of Intermolecular Interactions

Contact Info

Product

Resources

About