qNEP: A Highly Efficient Neuroevolution Potential with Dynamic Charges for Large-Scale Atomistic Simulations.

Overview

1. College of Physical Science and Technology, Bohai University, Jinzhou 121013, P. R. China
2. Suzhou Laboratory, Suzhou, Jiangsu 215123, P. R. China
3. Department of Physics, Chalmers University of Technology, 41296 Gothenburg, Sweden
4. Department of Materials Science and Engineering, Westlake University, Hangzhou, Zhejiang 310030, P. R. China
5. School of Science, Harbin Institute of Technology, Shenzhen, Guangdong 518055, P. R. China
6. Shenzhen Key Laboratory of Micro/Nano-Porous Functional Materials (SKLPM), Department of Materials Science and Engineering, Southern University of Science and Technology, Shenzhen 518055, P. R. China
7. Department of Materials Science and Engineering, City University of Hong Kong, Hong Kong SAR 999077, P. R. China
8. Department of Civil and Environmental Engineering, George Washington University, Washington, District of Columbia 20052, United States
9. National Engineering Laboratory for Reducing Emissions from Coal Combustion, Shandong Key Laboratory of Green Thermal Power and Carbon Reduction, Shandong University, Jinan, Shandong 250061, P. R. China
10. Wallenberg Initiative Materials Science for Sustainability, Chalmers University of Technology, 41926 Gothenburg, Sweden

Abstract

Although electrostatics can be incorporated into machine-learned interatomic potentials, existing approaches are computationally very demanding, limiting large-scale, long-time simulations of electrostatics-driven phenomena such as dielectric response, infrared activity, and field–matter coupling. Here, we extend the neuroevolution potential (NEP), a highly efficient machine-learned interatomic potential, to a charge-aware framework (qNEP) by introducing explicit, environment-dependent partial charges. Each ionic partial charge is represented by a neural network as a function of the local descriptor vector, analogous to the NEP site-energy model. This formulation enables the direct prediction of the Born effective charge tensor for each ion and, consequently, the polarization. As a result, dielectric properties, infrared spectra, and coupling to external electric fields can be evaluated within a unified framework. We derive consistent expressions for the forces and virials that explicitly account for the position dependence of the partial charges. The qNEP method has been implemented in the free-and-open-source GPUMD package with support for both Ewald summation and particle–particle particle–mesh treatments of electrostatics. We demonstrate the accuracy and efficiency of the qNEP approach through representative applications to water, Li7La3Zr2O12, BaTiO3, and a magnesium–water interface. These results show that qNEP enables accurate atomistic simulations with explicit long-range electrostatics, scalable to million-atom systems on nanosecond time scales using consumer-grade GPUs.

Reproduced under the paper's license (CC BY), from the paper cited above.

Code

The paper links to its data, not to its authors' code: see the Data section.

The paper's code and data availability statement is in the Data section.

Tracing map

A tracing map links a paper to the code its authors published: this paper has none, so it has no map.

Data

Datasets cited

• zenodo:18335946, at Zenodo; found in the availability statement
• zenodo:18335947, at Zenodo; found in the availability statement

Data availability

The NEP and qNEP models as well as the reference data used for their training and validation have been deposited on Zenodo under the Accession Code 10.5281/zenodo.18335946 (https://doi.org/10.5281/zenodo.18335947).

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 29 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 16 authors, 8 funders, 77 references.

Cite

This paper

Fan, Z., Tang, B., Berger, E., Berger, E., Fransson, E., Xu, K., Yan, Z., Liu, Z., Song, Z., Dong, H., Chen, S., Li, L., Wang, Z., Zhu, Y., Wiktor, J., & Erhart, P. (2026). qNEP: A Highly Efficient Neuroevolution Potential with Dynamic Charges for Large-Scale Atomistic Simulations. Journal of chemical theory and computation, 22(9), 4787-4801. https://doi.org/10.1021/acs.jctc.6c00146

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1073/pnas.2610049123 [code]

A unified machine-learning framework for ab initio multiscale modeling of liquids.

Journal: Proceedings of the National Academy of Sciences of the United States of America

In common: computational, 8 references

[2] doi:10.1085/jgp.202513926

Characterization of an open-channel structure and lateral conduction pathway in the cation-selective pentameric ligand-gated ion channel, ELIC.

Journal: The Journal of general physiology

In common: none (in silico), computational modeling (no new data), 1 reference

[3] doi:10.1038/s41467-026-74888-y

Engineered zwitterion-nanodelivery for precision targeting of brain metastases.

Journal: Nature communications

In common: 2 references

[4] doi:10.1021/acsomega.5c13203

Investigating the Function of C‑Terminal Tails of Human Tubulin Isotypes in the Motility Regulation of Cytoplasmic Dynein.

Journal: ACS omega

In common: 2 references

[5] doi: [code]

Going deeper with morphologically detailed neural networks by simulation-based gradient propagation

Journal: Frontiers in computational neuroscience

In common: none (in silico), computational, computational modeling (no new data)

[6] doi:10.1371/journal.pcbi.1014617 [code]

An in silico framework for dissecting the mechanistic origins of in vivo recorded neuronal activity.

Journal: PLoS computational biology

In common: none (in silico), computational, computational modeling (no new data)

[7] doi:10.1007/s11571-026-10522-3 [code]

Acetylcholine enhances deviance detection in Hodgkin-Huxley neuronal networks.

Journal: Cognitive neurodynamics

In common: none (in silico), computational, computational modeling (no new data)

[8] doi:10.1038/s41526-026-00644-7 [code]

A computational model of altered neuronal activity in altered gravity.

Journal: NPJ microgravity

In common: none (in silico), computational, computational modeling (no new data)

[9] doi:10.3389/fncom.2026.1799705

Structural synaptogenesis superior to functional modulation in a pruning-based recurrent network model of OCD.

Journal: Frontiers in computational neuroscience

In common: none (in silico), computational, computational modeling (no new data)

[10] doi:10.1371/journal.pcbi.1014458 [code]

Neuronal excitability and parameter variability in the Hodgkin-Huxley model.

Journal: PLoS computational biology

In common: none (in silico), computational, computational modeling (no new data)

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.