EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Cartesian equivariant representations for learning and understanding molecular orbitals

Daniel S King, Daniel Grzenda, Ray Zhu, Nathaniel Hudson, Ian Foster, Bingqing Cheng iD, Laura Gagliardi

Proceedings of the National Academy of Sciences · 02.12.2025 · 122 (48)

DOI: 10.1073/pnas.2510235122 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, online erschienen am 21.11.2025. Verfasst von Daniel S King und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemistry, University of Chicago, Chicago, IL 60637. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 84 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: US Department of energy office of science basic enery sciences.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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.

Schlagwörter

Stichwörter der Autor:innen

chemical reactionselectronic structuremachine learningmolecular orbitals

Autor:innen und Institutionen

  • Department of Chemistry, University of Chicago, Chicago, IL 60637.
  • Department of Computer Science, University of Chicago, Chicago, IL 60637.
  • Pritzker School of Molecular Engineering, Chicago, IL 60637.
  • Data Science and Learning Division, Argonne National Laboratory, Lemont, IL 60439.
  • Department of Chemistry, University of California, Berkeley, CA 94720.
  • The Institute of Science and Technology Austria, Klosterneuburg 3400, Austria.

ORCID-Profile: Bingqing Cheng

Förderung

  • US Department of energy office of science basic enery sciences

Zitieren

APA

King, D. S., 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, 122(48). https://doi.org/10.1073/pnas.2510235122

Vancouver

King DS, Grzenda D, Zhu R, Hudson N, Foster I, Cheng B, et al. Cartesian equivariant representations for learning and understanding molecular orbitals. Proceedings of the National Academy of Sciences. 2025;122(48). doi:10.1073/pnas.2510235122

BibTeX
@article{King2025,
  author = {King, Daniel S and Grzenda, Daniel and Zhu, Ray and Hudson, Nathaniel and Foster, Ian and Cheng, Bingqing and Gagliardi, Laura},
  title = {Cartesian equivariant representations for learning and understanding molecular orbitals},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2025},
  volume = {122},
  number = {48},
  doi = {10.1073/pnas.2510235122}
}
Kostenlos starten →