{"intvolume":"        17","oa":1,"OA_place":"publisher","language":[{"iso":"eng"}],"year":"2026","acknowledgement":"P.T. thanks valued discussions with Dr. Peichen Zhong, Dr. Hao Tang, and Dr. Chengbin Zhao. P.T. thanks Dr. Dingshun Lv, Dr. Zechang Sun, and Dr. Chenxi Hu for identifying an important bug in an early version of the code package. The authors acknowledge the resources of the National Energy Research Scientific Computing Center (NERSC), a Department of Energy Office of Science User Facility using NERSC award DOEERCAP0031751 ‘GenAI@NERSC’. P.T. acknowledges funding from the BIDMaP Postdoctoral Fellowship.","day":"21","arxiv":1,"PlanS_conform":"1","publication":"Nature Communications","volume":17,"article_type":"original","author":[{"full_name":"Tuo, Ping","last_name":"Tuo","id":"6e5644c0-c180-11ed-a2da-facc4c9f4f09","first_name":"Ping"},{"last_name":"Chen","full_name":"Chen, Jiale","id":"4d0a9064-1ff6-11ee-9fa6-ec046c604785","first_name":"Jiale","orcid":"0000-0001-5337-5875"},{"full_name":"Li, Ju","last_name":"Li","first_name":"Ju"}],"DOAJ_listed":"1","title":"Flow matching for reaction pathway generation","quality_controlled":"1","file":[{"relation":"main_file","creator":"dernst","access_level":"open_access","file_id":"22864","checksum":"3cfb714304eec29b97dcaf7fb86dd6e3","success":1,"file_size":1526298,"date_created":"2026-09-09T07:25:19Z","date_updated":"2026-09-09T07:25:19Z","file_name":"2026_NatureComm_Tuo.pdf","content_type":"application/pdf"}],"scopus_import":"1","publication_identifier":{"eissn":["2041-1723"]},"publisher":"Springer Nature","supplementarymaterial":"yes","file_date_updated":"2026-09-09T07:25:19Z","date_created":"2026-08-30T22:01:43Z","_id":"22771","department":[{"_id":"DaAl"},{"_id":"GradSch"}],"article_processing_charge":"Yes","has_accepted_license":"1","article_number":"8769","pmid":1,"license":"https://creativecommons.org/licenses/by/4.0/","publication_status":"published","date_updated":"2026-09-09T07:28:15Z","fulldoi":"https://doi.org/10.1038/s41467-026-75654-w","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","researchdata_availability":"yes","external_id":{"arxiv":["2507.10530"],"pmid":["42469237"]},"type":"journal_article","oa_version":"Published Version","das_tickbox":"1","abstract":[{"text":"Elucidating reaction mechanisms requires efficient generation of transition states (TSs) and products. Existing diffusion and sequence-based models accelerate parts of this process over traditional string-based methods, but typically still require manual enumeration of either TSs or products, and stochastic diffusion dynamics can be inefficient and hard to control. We introduce MolGEN, a conditional flow-matching framework that uses deterministic optimal transport to map Gaussian priors to chemical distributions. For TS generation, MolGEN improves TS geometry and barrier-height prediction over diffusion models while enabling sub-second sampling. For reaction product generation, it achieves competitive top-k accuracy while preserving mass and electron balance. Using the same backbone for TS and product sampling, MolGEN enables template-free generative exploration of reaction networks without the repeated quantum-chemistry searches required by prior methods. For the γ-ketohydroperoxide decomposition network, it produces more valid TSs than string-based methods using only 12 quantum-chemistry evaluations instead of 1156, and identifies a lower-barrier pathway.","lang":"eng"}],"month":"08","dataavailabilitystatement":"Structures generated in this study are provided in the Source Data file, and deposited in the GitHub repository https://github.com/tuoping/MolGEN and the figshare database under accession code https://doi.org/10.6084/m9.figshare.30576365. Source data are provided in this paper. The MolGEN codebase is available as an open-source repository for continuous development at https://github.com/tuoping/MolGEN. A release of the code used in this work has been archived on Zenodo47.","OA_type":"gold","doi":"10.1038/s41467-026-75654-w","status":"public","citation":{"chicago":"Tuo, Ping, Jiale Chen, and Ju Li. “Flow Matching for Reaction Pathway Generation.” <i>Nature Communications</i>. Springer Nature, 2026. <a href=\"https://doi.org/10.1038/s41467-026-75654-w\">https://doi.org/10.1038/s41467-026-75654-w</a>.","ama":"Tuo P, Chen J, Li J. Flow matching for reaction pathway generation. <i>Nature Communications</i>. 2026;17. doi:<a href=\"https://doi.org/10.1038/s41467-026-75654-w\">10.1038/s41467-026-75654-w</a>","ista":"Tuo P, Chen J, Li J. 2026. Flow matching for reaction pathway generation. Nature Communications. 17, 8769.","short":"P. Tuo, J. Chen, J. Li, Nature Communications 17 (2026).","ieee":"P. Tuo, J. Chen, and J. Li, “Flow matching for reaction pathway generation,” <i>Nature Communications</i>, vol. 17. Springer Nature, 2026.","mla":"Tuo, Ping, et al. “Flow Matching for Reaction Pathway Generation.” <i>Nature Communications</i>, vol. 17, 8769, Springer Nature, 2026, doi:<a href=\"https://doi.org/10.1038/s41467-026-75654-w\">10.1038/s41467-026-75654-w</a>.","apa":"Tuo, P., Chen, J., &#38; Li, J. (2026). Flow matching for reaction pathway generation. <i>Nature Communications</i>. Springer Nature. <a href=\"https://doi.org/10.1038/s41467-026-75654-w\">https://doi.org/10.1038/s41467-026-75654-w</a>"},"ddc":["000"],"date_published":"2026-08-21T00:00:00Z","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)","image":"/images/cc_by.png"}}