Molecular Dynamics of Alanine Dipeptide, Alanine Tetrapeptide, and Alanine Hexapeptide in Implicit Solvent

This repository contains the datasets used in our publication "Temperature-Annealed Boltzmann Generators" (https://arxiv.org/abs/2501.19077). See also our GitHub code repository that uses the provided datasets to evaluate trained Boltzmann generators (https://github.com/aimat-lab/TA-BG). We provide training, evaluation, and test datasets subsampled from molecular dynamics trajectories performed in OpenMM with Amber force fields and implicit solvation. We cover the molecular systems alanine dipeptide, alanine tetrapeptide, and alanine hexapeptide. Coordinates are given in Cartesian coordinates in angstroms. Details on simulation and force field parameters can be found in our publication. Overview of the folder structure and files contained in datasets.zip: aldp - Files for the alanine dipeptide system 300K_test_fab.npy - Test dataset at 300K, originally used in https://arxiv.org/abs/2208.01893 (Source: https://zenodo.org/records/6993124) 300K_train.npy - Training dataset at 300K, used for forward KLD experiments in our publication. 300K_val.npy - Validation dataset at 300K 1200K_val.npy - Validation dataset at 1200K tetra - Files for the alanine tetrapeptide system 300K_test.npy - Test dataset at 300K 300K_train.npy - Training dataset at 300K, used for forward KLD experiments in our publication. 300K_val.npy - Validation dataset at 300K 1200K_val.npy - Validation dataset at 1200K hexa - Files for the alanine hexapeptide system 300K_test.npy - Test dataset at 300K 300K_train.npy - Training dataset at 300K, used for forward KLD experiments in our publication. 300K_val.npy - Validation dataset at 300K 1200K_val.npy - Validation dataset at 1200K