Contrastive Domain Generalization for Cross-Instrument Molecular Identification in Mass Spectrometry

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Yoo, Seunghyun, Kim, Sanghong, Yoon, Namkyung, Kim, Hwangnam
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911413009645568
author Yoo, Seunghyun
Kim, Sanghong
Yoon, Namkyung
Kim, Hwangnam
author_facet Yoo, Seunghyun
Kim, Sanghong
Yoon, Namkyung
Kim, Hwangnam
contents Identifying molecules from mass spectrometry (MS) data remains a fundamental challenge due to the semantic gap between physical spectral peaks and underlying chemical structures. Existing deep learning approaches often treat spectral matching as a closed-set recognition task, limiting their ability to generalize to unseen molecular scaffolds. To overcome this limitation, we propose a cross-modal alignment framework that directly maps mass spectra into the chemically meaningful molecular structure embedding space of a pretrained chemical language model. On a strict scaffold-disjoint benchmark, our model achieves a Top-1 accuracy of 42.2% in fixed 256-way zero-shot retrieval and demonstrates strong generalization under a global retrieval setting. Moreover, the learned embedding space demonstrates strong chemical coherence, reaching 95.4% accuracy in 5-way 5-shot molecular re-identification. These results suggest that explicitly integrating physical spectral resolution with molecular structure embedding is key to solving the generalization bottleneck in molecular identification from MS data.
format Preprint
id arxiv_https___arxiv_org_abs_2602_00547
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Contrastive Domain Generalization for Cross-Instrument Molecular Identification in Mass Spectrometry
Yoo, Seunghyun
Kim, Sanghong
Yoon, Namkyung
Kim, Hwangnam
Machine Learning
Artificial Intelligence
I.2.1
Identifying molecules from mass spectrometry (MS) data remains a fundamental challenge due to the semantic gap between physical spectral peaks and underlying chemical structures. Existing deep learning approaches often treat spectral matching as a closed-set recognition task, limiting their ability to generalize to unseen molecular scaffolds. To overcome this limitation, we propose a cross-modal alignment framework that directly maps mass spectra into the chemically meaningful molecular structure embedding space of a pretrained chemical language model. On a strict scaffold-disjoint benchmark, our model achieves a Top-1 accuracy of 42.2% in fixed 256-way zero-shot retrieval and demonstrates strong generalization under a global retrieval setting. Moreover, the learned embedding space demonstrates strong chemical coherence, reaching 95.4% accuracy in 5-way 5-shot molecular re-identification. These results suggest that explicitly integrating physical spectral resolution with molecular structure embedding is key to solving the generalization bottleneck in molecular identification from MS data.
title Contrastive Domain Generalization for Cross-Instrument Molecular Identification in Mass Spectrometry
topic Machine Learning
Artificial Intelligence
I.2.1
url https://arxiv.org/abs/2602.00547