Transition1x: a dataset for building generalizable reactive machine learning potentials
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Transition1x : a dataset for building generalizable reactive machine learning potentials. / Schreiner, Mathias; Bhowmik, Arghya; Vegge, Tejs; Busk, Jonas; Winther, Ole.
In: Scientific Data, Vol. 9, No. 1, 779, 2022.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Transition1x
T2 - a dataset for building generalizable reactive machine learning potentials
AU - Schreiner, Mathias
AU - Bhowmik, Arghya
AU - Vegge, Tejs
AU - Busk, Jonas
AU - Winther, Ole
N1 - Publisher Copyright: © 2022, The Author(s).
PY - 2022
Y1 - 2022
N2 - Machine Learning (ML) models have, in contrast to their usefulness in molecular dynamics studies, had limited success as surrogate potentials for reaction barrier search. This is primarily because available datasets for training ML models on small molecular systems almost exclusively contain configurations at or near equilibrium. In this work, we present the dataset Transition1x containing 9.6 million Density Functional Theory (DFT) calculations of forces and energies of molecular configurations on and around reaction pathways at the ωB97x/6–31 G(d) level of theory. The data was generated by running Nudged Elastic Band (NEB) with DFT on 10k organic reactions of various types while saving intermediate calculations. We train equivariant graph message-passing neural network models on Transition1x and cross-validate on the popular ANI1x and QM9 datasets. We show that ML models cannot learn features in transition state regions solely by training on hitherto popular benchmark datasets. Transition1x is a new challenging benchmark that will provide an important step towards developing next-generation ML force fields that also work far away from equilibrium configurations and reactive systems.
AB - Machine Learning (ML) models have, in contrast to their usefulness in molecular dynamics studies, had limited success as surrogate potentials for reaction barrier search. This is primarily because available datasets for training ML models on small molecular systems almost exclusively contain configurations at or near equilibrium. In this work, we present the dataset Transition1x containing 9.6 million Density Functional Theory (DFT) calculations of forces and energies of molecular configurations on and around reaction pathways at the ωB97x/6–31 G(d) level of theory. The data was generated by running Nudged Elastic Band (NEB) with DFT on 10k organic reactions of various types while saving intermediate calculations. We train equivariant graph message-passing neural network models on Transition1x and cross-validate on the popular ANI1x and QM9 datasets. We show that ML models cannot learn features in transition state regions solely by training on hitherto popular benchmark datasets. Transition1x is a new challenging benchmark that will provide an important step towards developing next-generation ML force fields that also work far away from equilibrium configurations and reactive systems.
U2 - 10.1038/s41597-022-01870-w
DO - 10.1038/s41597-022-01870-w
M3 - Journal article
C2 - 36566281
AN - SCOPUS:85144636157
VL - 9
JO - Scientific data
JF - Scientific data
SN - 2052-4463
IS - 1
M1 - 779
ER -
ID: 330884489