Instrument-Agnostic Machine Learning Framework for Accurate Prediction of Functional Groups from Tandem Mass Spectrometry
Analytical Chemistry · 10.11.2026 · 98 (44) · 30122-30131
DOI: 10.1021/acs.analchem.5c05945 ↗Kurz gefasst
Fachartikel in der Zeitschrift Analytical Chemistry, online erschienen am 18.10.2026. Verfasst von Connor Beveridge und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, Indiana47907, United States. Die Arbeit nennt 26 Literaturangaben. Gefördert von: ACS Division of Analytical Chemistry, National Institutes of Health (NIH), NCATS NIH HHS.
Abstract (Originalsprache)
Structural elucidation of unknown compounds using tandem mass spectrometry (MS/MS) is an ongoing challenge. Expert chemists will often use common product ion peaks and neutral losses or peak matching software to predict structural aspects from MS/MS spectra, a process which is time consuming and limited by the molecular entries in the database. We introduce an instrument-agnostic machine learning (ML) framework that predicts functional groups directly from MS/MS spectra, eliminating the need for conventional database searches. The ML models are trained on data from different sources, the MassBank of North America (MONA), a public database, and a custom MS/MS database acquired using a single high throughput desorption electrospray ionization (DESI) MS/MS platform to showcase transferability of model predictions independent of instrument acquisition. Additionally, a novel spectral representation is developed for ML use by standardizing the spectra to the precursor ion m/z. We utilized molecular metrics, specifically, average molecular F1 score of 87% (MONA) and 76% (DESI) and average molecular accuracy of 94% (MONA) and 87% (DESI), indicating similar model performance when trained on different databases to validate the ML framework. To further illustrate the robustness and transferability of our ML framework, models trained on DESI-generated data successfully predicted functional groups, using spectral data generated from a blind test set of molecules using two different company laboratories using distinct mass spectrometry instruments (Orbitrap and TOF) operated in positive-ion mode electrospray ionization. We believe that such an approach, where models trained on data from one instrument can reliably make accurate predictions on spectra acquired using different instruments in a database-free manner, will significantly expand the practical applications of ML to mass spectrometry.
Autor:innen und Institutionen
- Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, Indiana47907, United States.
- Lilly Oncology @ Lilly, Lilly Oncology, Eli Lilly & Company, 600 Technology Ct., Louisville, Colorado80027, United States.
- Merck & Company, Rahway, New Jersey07033, United States.
- Corteva Agriscience, Indianapolis, Indiana46268, United States.
- Indiana Consortium for Analytical Science and Engineering (ICASE), Indianapolis, Indiana46202, United States.
- Agilent Technologies Inc. , Santa Clara, California95051, United States.
ORCID-Profile: Prageeth R Wijewardhane, Nicolás M Morato, Li-Kang Zhang, Douglas Richardson, Gaurav Chopra
Förderung
- ACS Division of Analytical Chemistry
- National Institutes of Health (NIH)
- NCATS NIH HHS
- National Science Foundation (NSF)
- National Center for Advancing Translational Sciences
Zitieren
APA
Beveridge, C., Wijewardhane, P. R., Muhoberac, M., Morato, N. M., Kaerner, A., Zhang, L. K., Richardson, D., Shoulds, A., Gilbert, J. R., Welch, C. J., Tichy, S., & Chopra, G. (2026). Instrument-Agnostic Machine Learning Framework for Accurate Prediction of Functional Groups from Tandem Mass Spectrometry. Analytical Chemistry, 98(44), 30122-30131. https://doi.org/10.1021/acs.analchem.5c05945
Vancouver
Beveridge C, Wijewardhane PR, Muhoberac M, Morato NM, Kaerner A, Zhang LK, et al. Instrument-Agnostic Machine Learning Framework for Accurate Prediction of Functional Groups from Tandem Mass Spectrometry. Analytical Chemistry. 2026;98(44):30122-30131. doi:10.1021/acs.analchem.5c05945
BibTeX
@article{Beveridge2026,
author = {Beveridge, Connor and Wijewardhane, Prageeth R and Muhoberac, Matthew and Morato, Nicolás M and Kaerner, Andreas and Zhang, Li-Kang and Richardson, Douglas and Shoulds, Ayanna and Gilbert, Jeffrey R and Welch, Christopher J and Tichy, Shane and Chopra, Gaurav},
title = {Instrument-Agnostic Machine Learning Framework for Accurate Prediction of Functional Groups from Tandem Mass Spectrometry},
journal = {Analytical Chemistry},
year = {2026},
volume = {98},
number = {44},
pages = {30122--30131},
doi = {10.1021/acs.analchem.5c05945}
}
EUVIMED