Transition state structure detection with machine learningś
Abstract Transition structure calculations via quantum chemistry methods have become a staple in modern chemical reaction research. Yet, success rates in optimizing transition structures rely heavily on rational initial guesses and expert supervision. We develop a machine learning approach that util...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | npj Computational Materials |
| Online Access: | https://doi.org/10.1038/s41524-025-01693-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849334454486564864 |
|---|---|
| author | Yitao Si Yiding Ma Tao Yu Yifan Wu Yingzhe Liu Weipeng Lai Zhixiang Zhang Jinwen Shi Liejin Guo Oleg V. Prezhdo Maochang Liu |
| author_facet | Yitao Si Yiding Ma Tao Yu Yifan Wu Yingzhe Liu Weipeng Lai Zhixiang Zhang Jinwen Shi Liejin Guo Oleg V. Prezhdo Maochang Liu |
| author_sort | Yitao Si |
| collection | DOAJ |
| description | Abstract Transition structure calculations via quantum chemistry methods have become a staple in modern chemical reaction research. Yet, success rates in optimizing transition structures rely heavily on rational initial guesses and expert supervision. We develop a machine learning approach that utilizes a bitmap representation of chemical structures to generate high-quality initial guesses for modeling transition states of chemical reactions. The core of the approach comprises a convolutional neural network methodology with a genetic algorithm. An extensive dataset derived from quantum chemistry computations is built, providing sufficient data on which the model can be trained, validated and tested. By applying the method to typical bi-molecular hydrogen abstraction reactions involving hydrofluorocarbons, hydrofluoroethers, and hydroxyl radicals—reactions critical in atmospheric fluoride degradation and global warming potential evaluation, yet extremely challenging to model, we achieve transition state optimizations with an impressive, verified success rate of 81.8% for hydrofluorocarbons and 80.9% for hydrofluoroethers. The reported work demonstrates the effectiveness of employing visual representation in chemical space exploration tasks and opens new avenues for the transition structure modeling. |
| format | Article |
| id | doaj-art-41b8d3dfdc4e4fdea10a4bac449f8e1a |
| institution | Kabale University |
| issn | 2057-3960 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Computational Materials |
| spelling | doaj-art-41b8d3dfdc4e4fdea10a4bac449f8e1a2025-08-20T03:45:34ZengNature Portfolionpj Computational Materials2057-39602025-07-0111111010.1038/s41524-025-01693-4Transition state structure detection with machine learningśYitao Si0Yiding Ma1Tao Yu2Yifan Wu3Yingzhe Liu4Weipeng Lai5Zhixiang Zhang6Jinwen Shi7Liejin Guo8Oleg V. Prezhdo9Maochang Liu10International Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow, Xi’an Jiaotong UniversityState Key Laboratory of Fluorine & Nitrogen Chemicals, Xi’an Modern Chemistry Research InstituteState Key Laboratory of Fluorine & Nitrogen Chemicals, Xi’an Modern Chemistry Research InstituteDepartments of Chemistry, and Physics and Astronomy, University of Southern CaliforniaState Key Laboratory of Fluorine & Nitrogen Chemicals, Xi’an Modern Chemistry Research InstituteState Key Laboratory of Fluorine & Nitrogen Chemicals, Xi’an Modern Chemistry Research InstituteState Key Laboratory of Fluorine & Nitrogen Chemicals, Xi’an Modern Chemistry Research InstituteInternational Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow, Xi’an Jiaotong UniversityInternational Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow, Xi’an Jiaotong UniversityDepartments of Chemistry, and Physics and Astronomy, University of Southern CaliforniaInternational Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow, Xi’an Jiaotong UniversityAbstract Transition structure calculations via quantum chemistry methods have become a staple in modern chemical reaction research. Yet, success rates in optimizing transition structures rely heavily on rational initial guesses and expert supervision. We develop a machine learning approach that utilizes a bitmap representation of chemical structures to generate high-quality initial guesses for modeling transition states of chemical reactions. The core of the approach comprises a convolutional neural network methodology with a genetic algorithm. An extensive dataset derived from quantum chemistry computations is built, providing sufficient data on which the model can be trained, validated and tested. By applying the method to typical bi-molecular hydrogen abstraction reactions involving hydrofluorocarbons, hydrofluoroethers, and hydroxyl radicals—reactions critical in atmospheric fluoride degradation and global warming potential evaluation, yet extremely challenging to model, we achieve transition state optimizations with an impressive, verified success rate of 81.8% for hydrofluorocarbons and 80.9% for hydrofluoroethers. The reported work demonstrates the effectiveness of employing visual representation in chemical space exploration tasks and opens new avenues for the transition structure modeling.https://doi.org/10.1038/s41524-025-01693-4 |
| spellingShingle | Yitao Si Yiding Ma Tao Yu Yifan Wu Yingzhe Liu Weipeng Lai Zhixiang Zhang Jinwen Shi Liejin Guo Oleg V. Prezhdo Maochang Liu Transition state structure detection with machine learningś npj Computational Materials |
| title | Transition state structure detection with machine learningś |
| title_full | Transition state structure detection with machine learningś |
| title_fullStr | Transition state structure detection with machine learningś |
| title_full_unstemmed | Transition state structure detection with machine learningś |
| title_short | Transition state structure detection with machine learningś |
| title_sort | transition state structure detection with machine learnings |
| url | https://doi.org/10.1038/s41524-025-01693-4 |
| work_keys_str_mv | AT yitaosi transitionstatestructuredetectionwithmachinelearnings AT yidingma transitionstatestructuredetectionwithmachinelearnings AT taoyu transitionstatestructuredetectionwithmachinelearnings AT yifanwu transitionstatestructuredetectionwithmachinelearnings AT yingzheliu transitionstatestructuredetectionwithmachinelearnings AT weipenglai transitionstatestructuredetectionwithmachinelearnings AT zhixiangzhang transitionstatestructuredetectionwithmachinelearnings AT jinwenshi transitionstatestructuredetectionwithmachinelearnings AT liejinguo transitionstatestructuredetectionwithmachinelearnings AT olegvprezhdo transitionstatestructuredetectionwithmachinelearnings AT maochangliu transitionstatestructuredetectionwithmachinelearnings |