Flexible Data Analysis Pipeline for High-Confidence Proteogenomics.

Area of Science:

• Genomics
• Proteomics
• Bioinformatics

Background:

• Proteogenomics enhances genome annotation by integrating proteomic data.
• Novel peptides offer direct evidence for unannotated protein-coding regions.
• Automated pipelines are crucial for high-throughput proteomic data analysis.

Purpose of the Study:

• To present a modular, automated pipeline for detecting novel peptides in proteomic datasets.
• To improve the accuracy and efficiency of identifying novel protein-coding regions.
• To facilitate proteogenomic analyses and enhance genome annotation.

Main Methods:

• Development of a modular, automated data analysis pipeline using the OpenMS framework.
• Incorporation of multiple database search engines for peptide identification.
• Application of machine learning (Percolator) for high-stringency peptide filtering and post-processing.

Main Results:

• Successful identification of novel peptides from proteomic data.
• Implementation of expert-defined criteria for high-stringency peptide identification.
• Demonstration of pipeline application on a human testis proteomic dataset.

Conclusions:

• The developed pipeline effectively identifies novel peptides, contributing to improved genome annotations.
• The pipeline led to the addition of five new gene annotations on the human reference genome.
• The proteogenomic approach provides valuable insights into unannotated genomic regions.