Efficient Prediction of Transition-Metal NMR Chemical Shifts Using Machine Learning: Do Two-Dimensional Descriptors Suffice?
Journal of Chemical Information and Modeling · 24.08.2026 · 66 (16) · 9916-9928
DOI: 10.1021/acs.jcim.6c01787 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, online erschienen am 18.06.2026. Verfasst von Yaroslav I Isaev und 4 weiteren Autor:innen. Beteiligte Einrichtung: G.A. Krestov Institute of Solution Chemistry of the Russian Academy of Sciences , 153045Ivanovo, Russian Federation. Hauptthemen laut MEDLINE: Machine Learning, Transition Elements. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 56 Literaturangaben. Gefördert von: Russian Science Foundation, Ministry of Science and Higher Education of the Russian Federation.
Abstract (Originalsprache)
Accurate prediction of nuclear magnetic resonance chemical shifts for transition-metal nuclei remains a challenging problem due to the high computational cost of quantum-chemical methods and the limited availability of experimental data. In this study, machine learning models were developed to predict chemical shifts of coordination compounds containing Mn, Fe, Nb, and Mo, using a curated data set of 1956 experimental measurements. Several approaches were evaluated, including descriptor-based models, graph neural networks, and transformer-based architectures. The Tabular Prior-Data Fitted Network model demonstrated the best performance across all nuclei, with prediction errors corresponding to 4.5–8% of the total chemical shift range. A comparison of molecular representations showed that two-dimensional descriptors provide accuracy comparable to three-dimensional approaches while requiring significantly lower computational cost. Model interpretation based on Shapley additive explanations revealed metal-specific structure–property relationships and enabled identification of the applicability domain. Attempts to construct unified models across different metals did not improve predictive performance, highlighting the element-specific nature of chemical shifts. External validation on an independent data set of 195Pt complexes confirmed the generalizability of our two-dimensional descriptor-based approach, achieving a holdout MAE of 159 ppm without any 3D structural information. These results demonstrate that machine learning models based on molecular descriptors provide an efficient and reliable alternative to quantum-chemical methods for predicting nuclear magnetic resonance chemical shifts of transition-metal compounds.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- G.A. Krestov Institute of Solution Chemistry of the Russian Academy of Sciences , 153045Ivanovo, Russian Federation.
- N.N. Semenov Federal Research Center for Chemical Physics of the Russian Academy of Sciences, 119991Moscow, Russian Federation.
ORCID-Profile: Alexander A Ksenofontov, Dmitriy M Makarov, Ilya A Khodov
Förderung
- Russian Science Foundation
- Ministry of Science and Higher Education of the Russian Federation
Zitieren
APA
Isaev, Y. I., Kovalev, A. E., Ksenofontov, A. A., Makarov, D. M., & Khodov, I. A. (2026). Efficient Prediction of Transition-Metal NMR Chemical Shifts Using Machine Learning: Do Two-Dimensional Descriptors Suffice?. Journal of Chemical Information and Modeling, 66(16), 9916-9928. https://doi.org/10.1021/acs.jcim.6c01787
Vancouver
Isaev YI, Kovalev AE, Ksenofontov AA, Makarov DM, Khodov IA. Efficient Prediction of Transition-Metal NMR Chemical Shifts Using Machine Learning: Do Two-Dimensional Descriptors Suffice?. Journal of Chemical Information and Modeling. 2026;66(16):9916-9928. doi:10.1021/acs.jcim.6c01787
BibTeX
@article{Isaev2026,
author = {Isaev, Yaroslav I and Kovalev, Alexey E and Ksenofontov, Alexander A and Makarov, Dmitriy M and Khodov, Ilya A},
title = {Efficient Prediction of Transition-Metal NMR Chemical Shifts Using Machine Learning: Do Two-Dimensional Descriptors Suffice?},
journal = {Journal of Chemical Information and Modeling},
year = {2026},
volume = {66},
number = {16},
pages = {9916--9928},
doi = {10.1021/acs.jcim.6c01787}
}
EUVIMED