EMLE-Engine¶

An engine for electrostatic embedding of machine learning potentials. Based on code by Kirill Zinovjev. Technical details can be found in our paper, available here. Please cite this work if you use emle-engine in your research. Supplementary information and data can be found here. For the original theory behind EMLE, please refer to this publication.

The general purpose embedding model currently supports HCNOS elements. We plan to add support for further elements in the near future.

We thank EPSRC for funding (grant code EP/V011421/1).

Citation¶

@article{doi:10.1021/acs.jctc.4c00248,
  author = {Zinovjev, Kirill and Hedges, Lester and Montagud Andreu, Rub{\'e}n and Woods, Christopher and Tu{\~n}ón, I{\~n}aki and van der Kamp, Marc W.},
  title = {emle-engine: A Flexible Electrostatic Machine Learning Embedding Package for Multiscale Molecular Dynamics Simulations},
  journal = {Journal of Chemical Theory and Computation},
  volume = {20},
  number = {11},
  pages = {4514-4522},
  year = {2024},
  doi = {10.1021/acs.jctc.4c00248},
  note ={PMID: 38804055},
  url = {https://doi.org/10.1021/acs.jctc.4c00248},
  eprint = {https://doi.org/10.1021/acs.jctc.4c00248}
}