tidy-MDQM9nc

This is a batteries-included version of the MDQM9nc dataset for training of Transferrable Samplers.

Overview

Split	HAC 1	HAC 2	HAC 3	HAC 4	HAC 5	HAC 6	HAC 7	HAC 8	HAC 9	Total mols	Conformers per mol
Train	3	5	7	23	63	218	673	2,412	8,902	12,306	16,000
Val	0	0	0	0	0	1	6	18	75	100	16,000
Test	0	0	0	0	0	1	6	18	75	100	36,000

• The training data were randomly sampled into a webdataset containing a (N,4) array of atomic coordinates, and a smiles txt file.
• The val/test data were kept in their original trajectory ordering, TICA models were fitted based on heavy atom pairwise distances, and stored as per-molecule npz-chunks.
• Topologies exist as tar archives for molecular mechanics potential energy functions, i.e. GAFF.

Citations

@dataset{viguera_diez_2024_10579242,
  author       = {Viguera Diez, Juan and
                  Olsson, Simon},
  title        = {MDQM9-nc dataset},
  month        = feb,
  year         = 2024,
  publisher    = {Zenodo},
  doi          = {10.26434/chemrxiv-2023-sx61w},
  url          = {https://doi.org/10.26434/chemrxiv-2023-sx61w},
}

@article{
doi:10.26434/chemrxiv-2023-sx61w,
author = {Juan Viguera Diez  and Sara Romeo Atance  and Ola Engkvist  and Simon Olsson },
title = {Generation of conformational ensembles of small molecules via Surrogate Model-Assisted Molecular Dynamics},
journal = {ChemRxiv},
volume = {2023},
number = {1124},
pages = {},
year = {2023},
doi = {10.26434/chemrxiv-2023-sx61w},
URL = {https://chemrxiv.org/doi/abs/10.26434/chemrxiv-2023-sx61w},
eprint = {https://chemrxiv.org/doi/pdf/10.26434/chemrxiv-2023-sx61w},
abstract = {The accurate prediction of thermodynamic properties is crucial in various fields such as drug discovery and materials design. This task relies on sampling from the underlying Boltzmann distribution, which is challenging using conventional approaches such as simulations. In this work, we introduce Surrogate Model-Assisted Molecular Dynamics (SMA-MD), a new procedure to sample the equilibrium ensemble of molecules. First, SMA-MD leverages Deep Generative Models to enhance the sampling of slow degrees of freedom.  Subsequently, the generated ensemble undergoes statistical reweighting, followed by short simulations. Our empirical results show that SMA-MD generates more diverse and lower energy ensembles than conventional Molecular Dynamics simulations. Furthermore, we showcase the application of SMA-MD for the computation of thermodynamical properties by estimating implicit solvation free energies.}}