A streamlined platform for analyzing tera-scale DDA and DIA mass spectrometry data enables highly sensitive...

Area of Science:

• Proteomics and Mass Spectrometry
• Bioinformatics and Computational Biology
• Immunology and Vaccinology

Background:

• Mass spectrometry, particularly data-dependent acquisition (DDA) and data-independent acquisition (DIA), is crucial for sensitive peptide analysis.
• Immunopeptidomics requires highly sensitive and specific methods for identifying disease-related peptides.
• Existing DDA and DIA data analysis platforms face limitations in sensitivity and specificity.

Purpose of the Study:

• To develop a streamlined, unified platform for analyzing both DDA and DIA mass spectrometry data.
• To enhance the sensitivity and specificity of peptide identification using deep learning.
• To apply the platform to immunopeptidomics for discovering potential vaccine targets.

Main Methods:

• Integration of deep learning-based spectral library search, database search, and de novo sequencing.
• Development of a unified computational framework for DDA and DIA data analysis.
• Validation on multiple benchmark and immunopeptidomics datasets.

Main Results:

• The platform identified 5-30% more peptide precursors compared to state-of-the-art systems.
• It identified 1.7-4.1 times more peptides from DDA data and 1.4-2.2 times more from DIA data in immunopeptidomics.
• Six T-cell epitopes from the SARS-CoV-2 immunopeptidome were discovered.

Conclusions:

• The developed platform significantly enhances peptide identification in mass spectrometry, especially for immunopeptidomics.
• It offers a sensitive and specific solution for analyzing large-scale datasets, supporting clinical applications.
• The identified T-cell epitopes represent potential targets for COVID-19 vaccine development.