Data and Software for "Beyond Multiscale: A Neural Network-Unified Molecular Model for Asymmetry-Free Dynamics Simulation"

<p>This repository contains the software patch, scripts, dataset, and pre-trained models for the Integrated Molecular Model proposed in the manuscript:</p> <p><strong>"Beyond Multiscale: A Neural Network-Unified Molecular Model for Asymmetry-Free Dynamics Simulation"</strong></p> <p>The provided files allow users to reproduce the active-learning protocol and execute stable, asymmetry-free multiscale molecular dynamics simulations.</p> <h3>Repository Structure:</h3> <ul> <li><strong>IMM_GROMACS.patch</strong> : The patch file containing the implementation of the Integrated Molecular Model for GROMACS.</li> <li><strong>pytorch_scripts/</strong> : PyTorch scripts used for the active-learning protocol and neural network training.</li> <li><strong>dataset/</strong> : Extracted data (weight functions and transition forces) and initial MD configurations required for training.</li> <li><strong>trained_models/</strong> : Pre-trained neural network weights for immediate MD simulations without retraining.</li> <li><strong>run_scripts/</strong> : Shell scripts to properly execute the uncorrected/corrected SCMP simulations and the training process.</li> <li><strong>environment.yml</strong> : Conda environment file to quickly reproduce the PyTorch and CUDA environment used in this study.</li> <li><strong>README.md</strong> : Detailed instruction file.</li> </ul><h3>License:</h3> <p>This repository contains both software code and data/models, which are distributed under different licenses:</p> <ul> <li><strong>IMM_GROMACS.patch</strong>: Licensed under the GNU Lesser General Public License v2.1 (LGPL-2.1), consistent with the base GROMACS 5.0.7 software.</li> <li><strong>Dataset, trained models, PyTorch scripts, and other files</strong>: Licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0).</li> </ul>