A Molecular Generation–Validation Workflow Combining Reversible Junction Tree Reinforcement Learning and Semi-empirical Quantum Chemistry: A Case Study on Molecular Solar Thermal Fuels
- 1Graduate School of Arts and Sciences, University of Tokyo, 3-8-1 Komaba, Meguro-ku, Tokyo, Japan, 153-8902
- 2Department of Chemistry, Advanced Materials Science and Engineering Center, and Institute for Energy Studies
- 3Western Washington University
Abstract
Molecular generative models can rapidly explore chemical space, but conventional metrics alone do not establish whether proposed structures possess application-relevant physical properties. To address this limitation, we developed an integrated generationvalidation framework coupling Reversible Junction Tree Reinforcement Learning (RJT- RL) with automated density-functional tight-binding (DFTB) evaluation. Azobenzenebased molecular solar thermal fuels served as a representative case study. RJT-RL was guided by maximum Tanimoto similarity to a reference dataset, while reference and generated molecules were evaluated under identical DFTB calculation conditions to obtain the energy difference between the cis and trans isomers (∆ E iso ) and the S 0 → S 1 excitation energy (∆ E ex ). As training progressed, the ∆ E iso distribution extended toward higher values, while ∆ E ex values were more concentrated within the visible-light window, yielding candidates within the selected joint screening region, although neither descriptor was included in the reward. These results show that automated semi-empirical post-generation validation can identify candidates with targeted property combinations and reveal information beyond conventional molecular generation metrics.
Related
- A Machine Learning Approach to logBB Prediction Combining Molecular Docking and Quantum-Chemical Descriptors
- Robust Multi-Objective Reinforcement Learning Improves Joint Predicted-Potency and Drug-Likeness Yield in EGFR Molecule Generation
- Design and computational evaluation of reversible covalent cathepsin S inhibitors using molecular docking, ADMET, molecular dynamics, MM/PBSA, and density functional theory
- Fully Reversible High-Capacity Ammonia Storage in a Series of Phase-Transformable Molecular Nickel Thiocarboxylates
- Azobenzene-Mediated Reversible Conformational Switching in 9,9′-Bifluorenylidene-Based Molecular Machines
BibTeX
@misc{zhu2026molecular,
title = {{A Molecular Generation–Validation Workflow Combining Reversible Junction Tree Reinforcement Learning and Semi-empirical Quantum Chemistry: A Case Study on Molecular Solar Thermal Fuels}},
author = {Qingyu Zhu and Ryan Lingg and Reuben Szabo and Tim Kowalczyk and Daisuke Yokogawa},
howpublished = {ChemRxiv},
year = {2026},
month = sep,
doi = {10.26434/chemrxiv.15009066/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009066/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - A Molecular Generation–Validation Workflow Combining Reversible Junction Tree Reinforcement Learning and Semi-empirical Quantum Chemistry: A Case Study on Molecular Solar Thermal Fuels
AU - Qingyu Zhu
AU - Ryan Lingg
AU - Reuben Szabo
AU - Tim Kowalczyk
AU - Daisuke Yokogawa
PY - 2026
DA - 2026/09/18
PB - ChemRxiv
DO - 10.26434/chemrxiv.15009066/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009066/v1
AB - Molecular generative models can rapidly explore chemical space, but conventional metrics alone do not establish whether proposed structures possess application-relevant physical properties. To address this limitation, we developed an integrated generationvalidation framework coupling Reversible Junction Tree Reinforcement Learning (RJT- RL) with automated density-functional tight-binding (DFTB) evaluation. Azobenzenebased molecular solar thermal fuels served as a representative case study. RJT-RL was guided by maximum Tanimoto similarity to a reference dataset, while reference and generated molecules were evaluated under identical DFTB calculation conditions to obtain the energy difference between the cis and trans isomers (∆ E iso ) and the S 0 → S 1 excitation energy (∆ E ex ). As training progressed, the ∆ E iso distribution extended toward higher values, while ∆ E ex values were more concentrated within the visible-light window, yielding candidates within the selected joint screening region, although neither descriptor was included in the reward. These results show that automated semi-empirical post-generation validation can identify candidates with targeted property combinations and reveal information beyond conventional molecular generation metrics.
ER -