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

Advancing Structure Elucidation with a Flexible Multi‐Spectral AI Model

Martin Priessner iD, Richard J Lewis iD, Isak Lemurell, Magnus J Johansson iD, JonathanM Goodman, Jon Paul Janet iD, Anna Tomberg iD

Angewandte Chemie International Edition · 09.01.2026 · 65 (2)

DOI: 10.1002/anie.202517611 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Angewandte Chemie International Edition, online erschienen am 26.11.2025. Verfasst von Martin Priessner und 6 weiteren Autor:innen. Beteiligte Einrichtung: Medicinal Chemistry, Research and Early Development, Cardiovascular, Renal and Metabolism, BioPharmaceuticals R&D, AstraZeneca, Pepparedsleden 1, Möl…. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Validating chemical synthesis success requires confirming the desired product using various analytical techniques. While spectroscopic data collection is increasingly automated, interpreting results remains a major bottleneck, often requiring expert input. With advances in laboratory automation and high‐throughput synthesis, this challenge is expected to intensify. We introduce the MultiModalSpectralTransformer (MMST), a machine learning method that predicts chemical structures directly from diverse spectral data (NMR, IR, and MS). Trained on 4 million simulated compounds, MMST achieves 72% and 80% as top‐1 and top‐3 accuracy, respectively. To address out‐of‐distribution challenges, we implemented an active learning improvement cycle that generates molecules in similar chemical spaces, enabling the model to adapt to chemical structures beyond its original training data. We demonstrate MMST's capabilities through comprehensive benchmarking across diverse molecular weight ranges and chemical spaces. Notably, despite training solely on simulated data, MMST demonstrates good performance with experimental spectra. This research represents a significant advancement in automated structure elucidation, offering a powerful and adaptable tool that bridges the gap between simulated and real‐world data.

Schlagwörter

Stichwörter der Autor:innen

Computer‐assisted structure elucidationIRMachine learningNMRTransformer

Autor:innen und Institutionen

  • Medicinal Chemistry, Research and Early Development, Cardiovascular, Renal and Metabolism, BioPharmaceuticals R&D, AstraZeneca, Pepparedsleden 1, Mölndal, 43183, Sweden.
  • Department of Medicinal Chemistry, Research & Early Development, Respiratory & Immunology, BioPharmaceuticals R&D, AstraZeneca, Pepparedsleden 1, Mölndal, 43183, Sweden.
  • Centre for Molecular Informatics, Yusuf Hamied Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, UK.
  • Molecular AI, Discovery Sciences, R&D, AstraZeneca, Pepparedsleden 1, Mölndal, 43183, Sweden.

ORCID-Profile: Martin Priessner, Richard J Lewis, Magnus J Johansson, Jon Paul Janet, Anna Tomberg

Zitieren

APA

Priessner, M., Lewis, R. J., Lemurell, I., Johansson, M. J., Goodman, J., Janet, J. P., & Tomberg, A. (2026). Advancing Structure Elucidation with a Flexible Multi‐Spectral AI Model. Angewandte Chemie International Edition, 65(2). https://doi.org/10.1002/anie.202517611

Vancouver

Priessner M, Lewis RJ, Lemurell I, Johansson MJ, Goodman J, Janet JP, et al. Advancing Structure Elucidation with a Flexible Multi‐Spectral AI Model. Angewandte Chemie International Edition. 2026;65(2). doi:10.1002/anie.202517611

BibTeX
@article{Priessner2026,
  author = {Priessner, Martin and Lewis, Richard J and Lemurell, Isak and Johansson, Magnus J and Goodman, JonathanM and Janet, Jon Paul and Tomberg, Anna},
  title = {Advancing Structure Elucidation with a Flexible Multi‐Spectral AI Model},
  journal = {Angewandte Chemie International Edition},
  year = {2026},
  volume = {65},
  number = {2},
  doi = {10.1002/anie.202517611}
}
Kostenlos starten →