MolecularDiffusion.modules.models.reactot¶
React-OT: a Schrodinger bridge that maps a reactant/product midpoint to the transition state in ten deterministic network evaluations.
Duan, C.; Liu, G.-H.; Du, Y.; …; Kulik, H. J., Optimal transport for generating transition states in chemical reactions, Nature Machine Intelligence (2025), doi:10.1038/s42256-025-01010-0. Preprint, under a different title: arXiv:2404.13430, React-OT: Optimal Transport for Generating Transition State in Chemical Reactions. Ported from https://github.com/deepprinciple/react-ot at commit 6dfccd0.
This package holds only what is new. The network is the same
object-aware LEFTNet the oareactdiff package already ships – verified
by diff -u: identical constructor, and a forward differing only by two
periodic-boundary kwargs whose pbc=False path is the code already here.
So there is deliberately no leftnet.py, dynamics.py,
normalizer.py or graph_tools.py; all four are imported from
modules/models/oareactdiff/, which is not edited. Cross-model imports
under modules/models/ are the in-tree pattern (ipdiff imports
kgdiff, flowmol_graph3d imports flowmol, geoldm imports
en_diffusion).
Its data layer is also shared: configs/data/reactot_ts1x_dataset.yaml
points at the existing OAReactDiffDataModule, and the task in
modules/tasks/diffusion_reactot.py imports _pad_ts from
modules/tasks/diffusion_oareactdiff.py.
See docs/model_integrations/reactot/INTEGRATION_PLAN.md.