AMP-BMS is a multiscale neural network potential trained on QM/MM data with electrostatic embedding for the simulations of biomolecules in the condensed phase. This repo was used for our recent publication together with a dataset for multiscale neural networks. This model is based on the AMP architecture and was used in previous work 1, 2.
Example usage is shown in here.
# Install environment.
conda env create -f amp.yml
conda activate amp_bms
# Install repo after cloning from github
pip install -e .
If you use this code, please cite our papers:
@article{AMP3,
title = {Multiscale Neural Network Potential with Anisotropic Message Passing for the Fast and Accurate Simulation of Protein Dynamics and Enzymatic Reactions},
volume = {148},
ISSN = {1520-5126},
url = {http://dx.doi.org/10.1021/jacs.6c00217},
DOI = {10.1021/jacs.6c00217},
number = {27},
journal = {J. Am. Chem. Soc.},
publisher = {American Chemical Society (ACS)},
author = {Th\"{u}rlemann, Moritz and Pultar, Felix and Gordiy, Igor and Ruijsenaars, Enrico and Riniker, Sereina},
year = {2026},
pages = {28133--28156}
}
@article{AMP2,
title = {Neural Network Potential with Multiresolution Approach Enables Accurate Prediction of Reaction Free Energies in Solution},
volume = {147},
ISSN = {1520-5126},
url = {http://dx.doi.org/10.1021/jacs.4c17015},
DOI = {10.1021/jacs.4c17015},
number = {8},
journal = {J. Am. Chem. Soc.},
publisher = {American Chemical Society (ACS)},
author = {Pultar, Felix and Th\"{u}rlemann, Moritz and Gordiy, Igor and Doloszeski, Eva and Riniker, Sereina},
year = {2025},
pages = {6835--6856}
}Moritz Thürlemann
Felix Pultar
Igor Gordiy
The AMP-BMS code is published and distributed under the MIT License.