ePrints

A Graph Neural Network Charge Model Targeting Accurate Electrostatic Properties of Organic Molecules

Lookup NU author(s): Dr Charlie Adams, Josh HortonORCiD, Dr Daniel ColeORCiD

Downloads

• Published version [.pdf]

Licence

This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).

Publication metadata

Author(s): Adams C, Horton JT, Wang L, Boothroyd S, Mobley DL, Wright DW, Cole DJ

Publication type: Article

Publication status: Published

Journal: Journal of Chemical Theory and Computation

Year: 2025

Volume: 21

Issue: 23

Pages: 12133-12148

Online publication date: 26/11/2025

Acceptance date: 18/11/2025

Date deposited: 04/12/2025

ISSN (print): 1549-9618

ISSN (electronic): 1549-9626

Publisher: American Chemical Society

URL: https://doi.org/10.1021/acs.jctc.5c01520

DOI: 10.1021/acs.jctc.5c01520

Data Access Statement: All models are openly available on GitHub, with a README that provides instructions for rapid charge assignment: https:// github.com/cole-group/nagl-mbis. The ChargeCraft psi4 wrapper is available at: https://github.com/bismuthadams1/ ChargeCraft. The data sets used to build our models are freely available on QCArchive: https://github.com/openforcefield/ qca-dataset-submission/tree/master/submissions/2024-07-26- MLPepper-RECAP-Optimized-Fragments-v1.0 and: https:// github.com/openforcefield/qca-dataset-submission/tree/ master/submissions/2024-10-11-MLPepper-RECAP- Optimized-Fragments-Add-Iodines-v1.0. The scripts associated with building, plotting, and processing the data sets to compare charge models can be found here: https://github.com/ bismuthadams1/charge_paper. LJ training workflows, includ- ing training data, scripts, and ForceBalance input files, are provided at: https://github.com/lilyminium/nagl-mbis-refit/.

Altmetrics

Funding

Share