FIORA: Local neighborhood-based prediction of compound mass spectra from single fragmentation events

Abstract Non-targeted metabolomics holds great promise for advancing precision medicine and biomarker discovery. However, identifying compounds from tandem mass spectra remains a challenging task due to the incomplete nature of spectral reference libraries. Augmenting these libraries with simulated...

Full description

Saved in:
Bibliographic Details
Main Authors: Yannek Nowatzky, Francesco Friedrich Russo, Jan Lisec, Alexander Kister, Knut Reinert, Thilo Muth, Philipp Benner
Format: Article
Language:English
Published: Nature Portfolio 2025-03-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-57422-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850072364010700800
author Yannek Nowatzky
Francesco Friedrich Russo
Jan Lisec
Alexander Kister
Knut Reinert
Thilo Muth
Philipp Benner
author_facet Yannek Nowatzky
Francesco Friedrich Russo
Jan Lisec
Alexander Kister
Knut Reinert
Thilo Muth
Philipp Benner
author_sort Yannek Nowatzky
collection DOAJ
description Abstract Non-targeted metabolomics holds great promise for advancing precision medicine and biomarker discovery. However, identifying compounds from tandem mass spectra remains a challenging task due to the incomplete nature of spectral reference libraries. Augmenting these libraries with simulated mass spectra can provide the necessary references to resolve unmatched spectra, but generating high-quality data is difficult. In this study, we present FIORA, an open-source graph neural network designed to simulate tandem mass spectra. Our main contribution lies in utilizing the molecular neighborhood of bonds to learn breaking patterns and derive fragment ion probabilities. FIORA not only surpasses state-of-the-art fragmentation algorithms, ICEBERG and CFM-ID, in prediction quality, but also facilitates the prediction of additional features, such as retention time and collision cross section. Utilizing GPU acceleration, FIORA enables rapid validation of putative compound annotations and large-scale expansion of spectral reference libraries with high-quality predictions.
format Article
id doaj-art-7c9eb1fa04064e9fa9b2dbf18ac4c636
institution DOAJ
issn 2041-1723
language English
publishDate 2025-03-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-7c9eb1fa04064e9fa9b2dbf18ac4c6362025-08-20T02:47:06ZengNature PortfolioNature Communications2041-17232025-03-0116111710.1038/s41467-025-57422-4FIORA: Local neighborhood-based prediction of compound mass spectra from single fragmentation eventsYannek Nowatzky0Francesco Friedrich Russo1Jan Lisec2Alexander Kister3Knut Reinert4Thilo Muth5Philipp Benner6Section VP.1 eScience, Federal Institute for Materials Research and Testing (BAM)Department of Analytical Chemistry and Reference Materials, Organic Trace Analysis and Food Analysis, Federal Institute for Materials Research and Testing (BAM)Department of Analytical Chemistry and Reference Materials, Organic Trace Analysis and Food Analysis, Federal Institute for Materials Research and Testing (BAM)Section VP.1 eScience, Federal Institute for Materials Research and Testing (BAM)Department of Mathematics and Computer Science, Freie Universität BerlinDepartment of Mathematics and Computer Science, Freie Universität BerlinSection VP.1 eScience, Federal Institute for Materials Research and Testing (BAM)Abstract Non-targeted metabolomics holds great promise for advancing precision medicine and biomarker discovery. However, identifying compounds from tandem mass spectra remains a challenging task due to the incomplete nature of spectral reference libraries. Augmenting these libraries with simulated mass spectra can provide the necessary references to resolve unmatched spectra, but generating high-quality data is difficult. In this study, we present FIORA, an open-source graph neural network designed to simulate tandem mass spectra. Our main contribution lies in utilizing the molecular neighborhood of bonds to learn breaking patterns and derive fragment ion probabilities. FIORA not only surpasses state-of-the-art fragmentation algorithms, ICEBERG and CFM-ID, in prediction quality, but also facilitates the prediction of additional features, such as retention time and collision cross section. Utilizing GPU acceleration, FIORA enables rapid validation of putative compound annotations and large-scale expansion of spectral reference libraries with high-quality predictions.https://doi.org/10.1038/s41467-025-57422-4
spellingShingle Yannek Nowatzky
Francesco Friedrich Russo
Jan Lisec
Alexander Kister
Knut Reinert
Thilo Muth
Philipp Benner
FIORA: Local neighborhood-based prediction of compound mass spectra from single fragmentation events
Nature Communications
title FIORA: Local neighborhood-based prediction of compound mass spectra from single fragmentation events
title_full FIORA: Local neighborhood-based prediction of compound mass spectra from single fragmentation events
title_fullStr FIORA: Local neighborhood-based prediction of compound mass spectra from single fragmentation events
title_full_unstemmed FIORA: Local neighborhood-based prediction of compound mass spectra from single fragmentation events
title_short FIORA: Local neighborhood-based prediction of compound mass spectra from single fragmentation events
title_sort fiora local neighborhood based prediction of compound mass spectra from single fragmentation events
url https://doi.org/10.1038/s41467-025-57422-4
work_keys_str_mv AT yanneknowatzky fioralocalneighborhoodbasedpredictionofcompoundmassspectrafromsinglefragmentationevents
AT francescofriedrichrusso fioralocalneighborhoodbasedpredictionofcompoundmassspectrafromsinglefragmentationevents
AT janlisec fioralocalneighborhoodbasedpredictionofcompoundmassspectrafromsinglefragmentationevents
AT alexanderkister fioralocalneighborhoodbasedpredictionofcompoundmassspectrafromsinglefragmentationevents
AT knutreinert fioralocalneighborhoodbasedpredictionofcompoundmassspectrafromsinglefragmentationevents
AT thilomuth fioralocalneighborhoodbasedpredictionofcompoundmassspectrafromsinglefragmentationevents
AT philippbenner fioralocalneighborhoodbasedpredictionofcompoundmassspectrafromsinglefragmentationevents