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

This repository contains the datasets used for training and evaluating Boltzmann generators. We provide training, validation, 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, alanine hexapeptide, and ELIL tetrapeptide, with details on force fields and data generation described in the publication "Learning Boltzmann Generators via Constrained Mass Transport" (https://arxiv.org/abs/2510.18460). Coordinates are given in Cartesian coordinates (Å). We only report data at 300K. Note: We observed small deviations from equilibrium in the energy histograms, so we slightly adjusted the MD procedure compared to previous publications (FAB, TA-BG) to generate conformations closer to the equilibrium distribution. Overview of the folder structure and files contained in datasets.zip: aldp – Files for the alanine dipeptide system 300K_test.npy – Test dataset at 300K 300K_train.npy – Training dataset at 300K 300K_val.npy – Validation dataset at 300K tetra – Files for the alanine tetrapeptide system 300K_test.npy – Test dataset at 300K 300K_train.npy – Training dataset at 300K 300K_val.npy – Validation dataset at 300K hexa – Files for the alanine hexapeptide system 300K_test.npy – Test dataset at 300K 300K_train.npy – Training dataset at 300K 300K_val.npy – Validation dataset at 300K ELIL tetrapeptide – Files for the ELIL tetrapeptide system 300K_test.npy – Test dataset at 300K 300K_train.npy – Training dataset at 300K 300K_val.npy – Validation dataset at 300K