Scholarlink

<p>Accurate identification of both glycan and peptide components is critical for mass spectrometry–based glycoproteomics, yet high-confidence glycan spectrum matches (GSMs) are often discarded during quality control due to low-confidence peptide spectrum matches (PSMs). To address this, we developed the&#xa0;glycopeptide boundary discriminator (GPBD), a new methodology that integrates glycan and peptide information to precisely define the glycan-peptide boundary. The GPBD framework consists of three main steps, for each spectrum: (1) enumerating several possible peptide masses from the core Y-ion ladder (low-energy scan) and then uses b/y evidence in the high-energy scan to rank peptide candidates, (2) for each candidate, matching a series of glycan ions to build a glycan ion pattern (GIP), (3) a classifier discriminates GIPs of the top candidates to finalize peptide assignment. This method is integrated into StrucGP, our previously developed software for N-glycopeptide identification. Experiments using GPBD on mouse brain and sperm datasets demonstrated an increase in glycopeptide spectra matches (GPSMs) due to more accurate peptide identification, while simultaneously maintaining high confidence in glycan identification. Furthermore, analysis of fetuin datasets showed GPBD’s potential usage in database-independent peptide identification pipeline.</p> Graphical abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

GPBD: high-confidence N-glycopeptide identification via glycopeptide boundary discriminator

  • 期刊论文

  • 发表日期：

    2025年11月1日

查看全文
  • Qiudi Ye,
  • Xiguo Yuan,
  • Shisheng Sun,
  • Xue Li

摘要

Accurate identification of both glycan and peptide components is critical for mass spectrometry–based glycoproteomics, yet high-confidence glycan spectrum matches (GSMs) are often discarded during quality control due to low-confidence peptide spectrum matches (PSMs). To address this, we developed the glycopeptide boundary discriminator (GPBD), a new methodology that integrates glycan and peptide information to precisely define the glycan-peptide boundary. The GPBD framework consists of three main steps, for each spectrum: (1) enumerating several possible peptide masses from the core Y-ion ladder (low-energy scan) and then uses b/y evidence in the high-energy scan to rank peptide candidates, (2) for each candidate, matching a series of glycan ions to build a glycan ion pattern (GIP), (3) a classifier discriminates GIPs of the top candidates to finalize peptide assignment. This method is integrated into StrucGP, our previously developed software for N-glycopeptide identification. Experiments using GPBD on mouse brain and sperm datasets demonstrated an increase in glycopeptide spectra matches (GPSMs) due to more accurate peptide identification, while simultaneously maintaining high confidence in glycan identification. Furthermore, analysis of fetuin datasets showed GPBD’s potential usage in database-independent peptide identification pipeline.

Graphical abstract