{"external_id":{"pmid":["41269783"]},"scopus_import":"1","day":"02","date_published":"2025-12-02T00:00:00Z","publication_status":"published","acknowledgement":"This work is supported as part of the Catalyst Design for Decarbonization Center, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under award no. DE-SC0023383. We thank the Research Computing Center at the University of Chicago and for access to computational resources. Additionally, this research used the Savio computational cluster resource provided by the Berkeley Research Computing program at the University of California (UC), Berkeley (supported by the UC Berkeley Chancellor, Vice Chancellor for Research, and Chief Information Officer). Furthermore, we thank Matthew Hennefarth and Matt Hermes for useful discussions.","abstract":[{"lang":"eng","text":"Qualitative and quantitative orbital properties such as bonding/antibonding character, localization, and orbital energies are critical to how chemists understand reactivity, catalysis, and excited-state behavior. Despite this, representations of orbitals in deep learning models have been very underdeveloped relative to representations of molecular geometries and Hamiltonians. Here, we apply state-of-the-art equivariant deep learning architectures to the task of assigning global labels to orbitals, namely energies characterizations, given the molecular coefficients from Hartree–Fock or density functional theory. The architecture we have developed, the Cartesian Equivariant Orbital Network (CEONET), shows how molecular orbital coefficients are readily featurized as equivariant node features common to all graph-based machine-learned potentials. We find that CEONET performs well at predicting difficult quantitative labels such as the orbital energy and orbital entropy. Furthermore, we find that the CEONET representation provides an intuitive latent space for differentiating orbital character for the qualitative assignment of e.g. bonding or antibonding character. In addition to providing a useful representation for further integrating deep learning with electronic structure theory, we expect CEONET to be useful for automatizing and interpreting the results of advanced electronic structure methods such as complete active space self-consistent field theory. In particular, the ability of CEONET to infer multireference character via the orbital entropy paves the way toward the machine-learned selection of active spaces."}],"license":"https://creativecommons.org/licenses/by-nc-nd/4.0/","department":[{"_id":"BiCh"}],"publication":"Proceedings of the National Academy of Sciences of the United States of America","language":[{"iso":"eng"}],"publication_identifier":{"eissn":["1091-6490"]},"oa":1,"OA_type":"hybrid","date_created":"2025-11-30T23:02:06Z","date_updated":"2025-12-01T09:02:29Z","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","OA_place":"publisher","_id":"20702","ddc":["540"],"pmid":1,"volume":122,"status":"public","type":"journal_article","month":"12","oa_version":"Published Version","has_accepted_license":"1","year":"2025","author":[{"full_name":"King, Daniel S.","last_name":"King","first_name":"Daniel S."},{"full_name":"Grzenda, Daniel","last_name":"Grzenda","first_name":"Daniel"},{"full_name":"Zhu, Ray","first_name":"Ray","last_name":"Zhu"},{"first_name":"Nathaniel","last_name":"Hudson","full_name":"Hudson, Nathaniel"},{"first_name":"Ian","last_name":"Foster","full_name":"Foster, Ian"},{"id":"cbe3cda4-d82c-11eb-8dc7-8ff94289fcc9","full_name":"Cheng, Bingqing","first_name":"Bingqing","orcid":"0000-0002-3584-9632","last_name":"Cheng"},{"last_name":"Gagliardi","first_name":"Laura","full_name":"Gagliardi, Laura"}],"issue":"48","quality_controlled":"1","article_type":"original","intvolume":"       122","article_number":"e2510235122","tmp":{"name":"Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","image":"/images/cc_by_nc_nd.png","short":"CC BY-NC-ND (4.0)","legal_code_url":"https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode"},"file":[{"access_level":"open_access","relation":"main_file","file_name":"2025_PNAS_King.pdf","checksum":"58051539a884c7a97306fd3afdb539ac","file_id":"20719","creator":"dernst","success":1,"content_type":"application/pdf","date_created":"2025-12-01T08:41:32Z","date_updated":"2025-12-01T08:41:32Z","file_size":27607870}],"corr_author":"1","doi":"10.1073/pnas.2510235122","article_processing_charge":"Yes (in subscription journal)","file_date_updated":"2025-12-01T08:41:32Z","related_material":{"link":[{"url":"https://github.com/GagliardiGroup/CEONet ","relation":"software"}]},"title":"Cartesian equivariant representations for learning and understanding molecular orbitals","citation":{"chicago":"King, Daniel S., Daniel Grzenda, Ray Zhu, Nathaniel Hudson, Ian Foster, Bingqing Cheng, and Laura Gagliardi. “Cartesian Equivariant Representations for Learning and Understanding Molecular Orbitals.” <i>Proceedings of the National Academy of Sciences of the United States of America</i>. National Academy of Sciences, 2025. <a href=\"https://doi.org/10.1073/pnas.2510235122\">https://doi.org/10.1073/pnas.2510235122</a>.","apa":"King, D. S., Grzenda, D., Zhu, R., Hudson, N., Foster, I., Cheng, B., &#38; Gagliardi, L. (2025). Cartesian equivariant representations for learning and understanding molecular orbitals. <i>Proceedings of the National Academy of Sciences of the United States of America</i>. National Academy of Sciences. <a href=\"https://doi.org/10.1073/pnas.2510235122\">https://doi.org/10.1073/pnas.2510235122</a>","mla":"King, Daniel S., et al. “Cartesian Equivariant Representations for Learning and Understanding Molecular Orbitals.” <i>Proceedings of the National Academy of Sciences of the United States of America</i>, vol. 122, no. 48, e2510235122, National Academy of Sciences, 2025, doi:<a href=\"https://doi.org/10.1073/pnas.2510235122\">10.1073/pnas.2510235122</a>.","ista":"King DS, Grzenda D, Zhu R, Hudson N, Foster I, Cheng B, Gagliardi L. 2025. Cartesian equivariant representations for learning and understanding molecular orbitals. Proceedings of the National Academy of Sciences of the United States of America. 122(48), e2510235122.","ieee":"D. S. King <i>et al.</i>, “Cartesian equivariant representations for learning and understanding molecular orbitals,” <i>Proceedings of the National Academy of Sciences of the United States of America</i>, vol. 122, no. 48. National Academy of Sciences, 2025.","short":"D.S. King, D. Grzenda, R. Zhu, N. Hudson, I. Foster, B. Cheng, L. Gagliardi, Proceedings of the National Academy of Sciences of the United States of America 122 (2025).","ama":"King DS, Grzenda D, Zhu R, et al. Cartesian equivariant representations for learning and understanding molecular orbitals. <i>Proceedings of the National Academy of Sciences of the United States of America</i>. 2025;122(48). doi:<a href=\"https://doi.org/10.1073/pnas.2510235122\">10.1073/pnas.2510235122</a>"},"publisher":"National Academy of Sciences"}