Pepitome: evaluating improved spectral library search for identification complementarity and quality assessment

Area of Science:

• Proteomics
• Bioinformatics
• Mass Spectrometry

Background:

• Spectral libraries offer an alternative to protein sequence databases for peptide identification.
• Current methods often use dot product scoring, which lacks statistical rigor and ignores ion m/z discrepancies.

Purpose of the Study:

• To develop a novel spectral library search engine, Pepitome, with improved scoring for spectrum-spectrum matches (SSMs).
• To evaluate Pepitome's performance against existing tools and assess its utility in proteomics workflows.

Main Methods:

• Developed Pepitome, a spectral library search engine utilizing statistical scoring for SSMs.
• Compared Pepitome to SpectraST on datasets from three mass spectrometry platforms.
• Validated spectral library search reliability using RNA-Seq data.

Main Results:

• Pepitome demonstrated superior performance compared to SpectraST across different mass spectrometry platforms.
• Spectral library searches were confirmed as reliable through RNA-Seq data validation.
• Pepitome facilitated automated quality analysis and control (QA/QC) in proteomics pipelines.

Conclusions:

• Pepitome offers a statistically robust and high-performing alternative for spectral library searches in peptide identification.
• Spectral library and database search methods are complementary.
• Pepitome enhances the automation and reliability of proteomics data acquisition and analysis.