Creative Commons source reading · Prediction and forecasting
Deep Learning Prediction of O‐Glycopeptide Tandem Mass Spectra Enhances O‐Glycoproteomics
Yu Zong; Yuxin Wang; Liang Qiao
Originally published in Adv Sci (Weinh) · 2026-07-23
From the article
Selected source text. JournalOS removes markup and shortens excerpts where indicated. The full article provides the complete context.
Abstract
ABSTRACTProtein glycosylation, a post‐translational modification involving the attachment of glycans to proteins, plays critical roles in numerous physiological and pathological cellular functions. Characterization of protein glycosylation is one of the most challenging problems due to the high heterogeneity of glycosites and glycan structures. Recently, deep learning has been adopted to predict N‐glycopeptide tandem mass spectrometry (MS/MS) spectra and exhibited a promising effect in N‐glycoproteomics analysis. However, current deep learning frameworks struggle to accurately predict O‐glycopeptide MS/MS spectra due to the complexity of O‐glycopeptides and the limited availability of training data. In this study, we introduce DeepGPO, a deep learning framework for the prediction of O‐glycopeptide MS/MS spectra. The DeepGPO…
Discussion
In this work, we present DeepGPO, a deep‐learning model capable of accurately predicting MS/MS spectra for O‐glycopeptides. Based on our previous DeepGP model for N‐glycopeptides, DeepGPO adopts a number of advanced training techniques to mitigate the limited availability of O‐glycopeptides training data. The introduction of a loss re‐weighting method preserves a wider range of MS/MS spectra for model training while assigning greater contributions to more reliable MS/MS spectra during training. Retaining replicate MS/MS spectra provides naturally occurring variation that functions as a form of data augmentation, where the intrinsic variations in different MS/MS spectra for the same glycopeptide are captured by the deep‐learning model. Since many fragmentation events in N‐glycopeptides can also happen in O‐glycopeptides, DeepGPO also…
Editorial reading guide
- What data created and tested the model, and was evaluation independent?
- What is the cost of a false positive, false negative or out-of-distribution decision?
