EUVIMEDEUVIMED entdecken
Crossref

Can Large Language Models Translate MS/MS into Molecular Caption?

Menglin Zhou, Chenhao Gong, Shan Cong, Qiong Yang, Siyu Jiang, Jing Tian, Changgen Xie, Sihao Chang, Hongchao Ji iD

Analytical Chemistry · 28.07.2026 · 98 (29) · 21148-21156

DOI: 10.1021/acs.analchem.6c02505 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Analytical Chemistry, online erschienen am 16.07.2026. Verfasst von Menglin Zhou und 8 weiteren Autor:innen. Beteiligte Einrichtung: Shenzhen Branch, Guangdong Laboratory for Lingnan Modern Agriculture, Genome Analysis Laboratory, Ministry of Agriculture and Rural Affairs, Agricult…. Hauptthemen laut MEDLINE: Large Language Models, Tandem Mass Spectrometry. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 20 Literaturangaben. Gefördert von: Harbin Engineering University, Agricultural Science and Technology Innovation Program, National Natural Science Foundation of China (NSFC).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Mass spectrometry (MS) is central to molecular discovery, yet the interpretation of tandem mass spectra (MS/MS) remains limited by database dependence and an incomplete structural resolution. Here, we explore whether large language models (LLMs) can directly translate MS/MS spectra to chemically meaningful molecular descriptions. We introduce MS2LLM, a framework that represents spectra and molecular structures as natural languages and learns their correspondence through instruction tuning. MS2LLM generates hierarchical molecular descriptions, including functional groups, substructures, and chemical classes, directly from the spectral input. Across multiple datasets, it outperforms general-purpose LLMs and conventional spectral learning methods in both descriptive accuracy and chemical classification. Importantly, the model produces interpretable outputs that capture structural semantics rather than exact structures, offering a complementary paradigm for structural inference of unknowns.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Large Language ModelsTandem Mass Spectrometry

Autor:innen und Institutionen

  • Shenzhen Branch, Guangdong Laboratory for Lingnan Modern Agriculture, Genome Analysis Laboratory, Ministry of Agriculture and Rural Affairs, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural S…
  • College of Life Science, Northwest A&F University, Yangling, Shaanxi712100, People's Republic of China.
  • Qingdao Innovation and Development Center, Harbin Engineering University, Qingdao266000, People's Republic of China.
  • School of Pharmacy, Hunan University of Chinese Medicine, Changsha, Hunan410208, People's Republic of China.

ORCID-Profile: Hongchao Ji

Förderung

  • Harbin Engineering University
  • Agricultural Science and Technology Innovation Program
  • National Natural Science Foundation of China (NSFC)
  • Science and Technology Innovation Program of the Chinese Academy of Agricultural Sciences

Zitieren

APA

Zhou, M., Gong, C., Cong, S., Yang, Q., Jiang, S., Tian, J., Xie, C., Chang, S., & Ji, H. (2026). Can Large Language Models Translate MS/MS into Molecular Caption?. Analytical Chemistry, 98(29), 21148-21156. https://doi.org/10.1021/acs.analchem.6c02505

Vancouver

Zhou M, Gong C, Cong S, Yang Q, Jiang S, Tian J, et al. Can Large Language Models Translate MS/MS into Molecular Caption?. Analytical Chemistry. 2026;98(29):21148-21156. doi:10.1021/acs.analchem.6c02505

BibTeX
@article{Zhou2026,
  author = {Zhou, Menglin and Gong, Chenhao and Cong, Shan and Yang, Qiong and Jiang, Siyu and Tian, Jing and Xie, Changgen and Chang, Sihao and Ji, Hongchao},
  title = {Can Large Language Models Translate MS/MS into Molecular Caption?},
  journal = {Analytical Chemistry},
  year = {2026},
  volume = {98},
  number = {29},
  pages = {21148--21156},
  doi = {10.1021/acs.analchem.6c02505}
}
Kostenlos starten →