Mass spectrum prediction improves with on-site reference-guided updates
Transferable Mass Spectrum Prediction via Reference-Guided Test-time Specialization
Machine Learning
Summary
Predicting the patterns of molecules breaking apart in mass spectrometry helps scientists identify chemicals. However, these predictions often get worse when the chemicals or testing conditions change. The authors introduce SPARC, a method that adapts existing predictors at the time of testing by using related reference data, without needing to see new test spectra first. This approach improves prediction accuracy for different chemical groups and testing setups, making it easier to use mass spectrometry in diverse scientific areas.
What this means in practice
- •For metabolomics labs: Improve chemical identification by adapting mass spectrum predictors to specific compound sets and lab conditions using onsite reference data.
- •For environmental monitoring teams: Enhance detection of environmental compounds by refining spectrum predictions according to local chemical variations without extensive retraining.
Authors
Yunhua Zhong, Runting Li, Yifan Li, Pan Liu, Zhiwen Yang, Zikun Wang, Yixuan Tang, Jun Xia
Abstract
Tandem mass spectrum prediction supports compound identification across metabolomics, natural-product discovery, and environmental analysis. However, pretrained predictors often degrade under shifts in chemical space and acquisition conditions, while retraining domain-specific models from scratch is costly. We introduce SPARC, a retrieval-guided test-time specialization framework that adapts a pretrained predictor using a spectral reference library without accessing test-query spectra. For each target query, SPARC retrieves chemically related reference spectra to recalibrate fragment intensities within the learned fragmentation space. During Transfer, SPARC combines reference-guided spectral adaptation with reliability-aware consistency, using reconstruction behavior on retrieved spectra to selectively preserve trustworthy predictions during continual specialization. Across MassSpecGym, NPLIB1 and application-specific GNPS libraries, SPARC improves spectral prediction under multiple transfer settings. These results establish retrieval-guided test-time specialization as a practical strategy for extending pretrained MS/MS predictors to specific chemical and acquisition domains, with continual test-time training providing further refinement during deployment.