Neural network interatomic potentials (NNPs) have recently proven to be powerful tools to accurately model complex molecular systems while bypassing the high numerical cost of ab initio molecular dynamics simulations. In recent years, numerous advances in model architectures as well as the development of hybrid models combining machine-learning (ML) with more traditional, physically motivated, force-field interactions have considerably increased the design space of ML potentials. In this paper, we present FeNNol, a new library for building, training, and running force-field-enhanced neural network potentials. It provides a flexible and modular system for building hybrid models, allowing us to easily combine state-of-the-art embeddings with ML-parameterized physical interaction terms without the need for explicit programming. Furthermore, FeNNol leverages the automatic differentiation and just-in-time compilation features of the Jax Python library to enable fast evaluation of NNPs, shrinking the performance gap between ML potentials and standard force-fields. This is demonstrated with the popular ANI-2x model reaching simulation speeds nearly on par with the AMOEBA polarizable force-field on commodity GPUs (graphics processing units). We hope that FeNNol will facilitate the development and application of new hybrid NNP architectures for a wide range of molecular simulation problems.
Skip Nav Destination
Article navigation
28 July 2024
Research Article|
July 25 2024
FeNNol: An efficient and flexible library for building force-field-enhanced neural network potentials Available to Purchase
Special Collection:
Modular and Interoperable Software for Chemical Physics
Thomas Plé
;
Thomas Plé
a)
(Conceptualization, Data curation, Software, Validation, Writing – original draft)
Sorbonne Université, LCT, UMR 7616 CNRS
, 75005 Paris, France
Search for other works by this author on:
Olivier Adjoua;
Olivier Adjoua
(Software, Writing – review & editing)
Sorbonne Université, LCT, UMR 7616 CNRS
, 75005 Paris, France
Search for other works by this author on:
Louis Lagardère
;
Louis Lagardère
(Software, Writing – review & editing)
Sorbonne Université, LCT, UMR 7616 CNRS
, 75005 Paris, France
Search for other works by this author on:
Jean-Philip Piquemal
Jean-Philip Piquemal
a)
(Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Writing – original draft, Writing – review & editing)
Sorbonne Université, LCT, UMR 7616 CNRS
, 75005 Paris, France
Search for other works by this author on:
Thomas Plé
Conceptualization, Data curation, Software, Validation, Writing – original draft
a)
Sorbonne Université, LCT, UMR 7616 CNRS
, 75005 Paris, France
Olivier Adjoua
Software, Writing – review & editing
Sorbonne Université, LCT, UMR 7616 CNRS
, 75005 Paris, France
Louis Lagardère
Software, Writing – review & editing
Sorbonne Université, LCT, UMR 7616 CNRS
, 75005 Paris, France
Jean-Philip Piquemal
Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Writing – original draft, Writing – review & editing
a)
Sorbonne Université, LCT, UMR 7616 CNRS
, 75005 Paris, France
J. Chem. Phys. 161, 042502 (2024)
Article history
Received:
May 06 2024
Accepted:
June 28 2024
Citation
Thomas Plé, Olivier Adjoua, Louis Lagardère, Jean-Philip Piquemal; FeNNol: An efficient and flexible library for building force-field-enhanced neural network potentials. J. Chem. Phys. 28 July 2024; 161 (4): 042502. https://doi.org/10.1063/5.0217688
Download citation file:
Pay-Per-View Access
$40.00
Sign In
You could not be signed in. Please check your credentials and make sure you have an active account and try again.
Citing articles via
CREST—A program for the exploration of low-energy molecular chemical space
Philipp Pracht, Stefan Grimme, et al.
DeePMD-kit v2: A software package for deep potential models
Jinzhe Zeng, Duo Zhang, et al.
Related Content
Considerations in the use of machine learning force fields for free energy calculations
J. Chem. Phys. (May 2025)
Machine learning-based dengue forecasting system for Irisan, Baguio city, Philippines
AIP Conf. Proc. (August 2022)
Applications of machine learning in surfaces and interfaces
Chem. Phys. Rev. (March 2025)