Integrated Proteomic Pipeline Using Multiple Search Engines for a Proteogenomic Study with a Controlled Protein False...

Area of Science:

• Proteomics
• Bioinformatics
• Genomics

Background:

• False-positive protein identifications are a significant challenge in large-scale proteomic studies, particularly within the Chromosome-Centric Human Proteome Project (C-HPP).
• Accurate protein identification is crucial for proteogenomic studies utilizing liquid-chromatography and mass-spectrometry-based profiling.

Purpose of the Study:

• To develop and validate a simple, integrated proteomic pipeline (IPP) for improved protein identification with controlled false discovery rates (FDR).
• To enhance the identification of alternative splicing variants and 'missing proteins' in human proteomic datasets.

Main Methods:

• Utilized three search engines (SEQUEST, MASCOT, MS-GF+) against the neXtProt database.
• Combined and statistically evaluated peptide spectrum matches (PSMs) using DTASelect and Percolator.
• Normalized peptide search scores to E-scores and employed ProteinInferencer for protein filtering at a 1.0% protein-level FDR.

Main Results:

• The IPP identified 5756 proteins from human hippocampal tissue, compared to 4453 using a conventional pipeline (CPP).
• Significantly more alternative splicing variants (477 vs. 182) and missing proteins (10 vs. 7) were identified using the IPP.
• Validated tryptic peptides for identified missing proteins using MS/MS spectral patterns or synthetic peptides.

Conclusions:

• The developed integrated proteomic pipeline (IPP) effectively improves protein identification accuracy and depth in human proteomic studies.
• The IPP is particularly beneficial for identifying challenging targets like alternative splicing variants and missing proteins for the C-HPP.
• This strategy provides a robust method for controlled protein identification at the proteome level.