Identifying PE2 and PE5 Proteins from Existing Mass Spectrometry Data Using pFind.

Area of Science:

• Proteomics
• Genomics
• Bioinformatics

Background:

• The Chromosome-Centric Human Proteome Project (C-HPP) aims to identify all human proteins.
• Approximately 2000 PE2-PE5 proteins lack sufficient protein-level evidence, posing a challenge for C-HPP.
• Previous C-HPP efforts faced difficulties in identifying these low-evidence proteins due to limited occurrences.

Purpose of the Study:

• To investigate if reanalyzing massive mass spectrometry (MS) data sets with novel algorithms can increase the identification of PE2-PE5 proteins.
• To provide evidence for the existence of PE2-PE5 proteins and assess their properties.
• To demonstrate the value of mining existing data repositories for discovering uncharacterized proteins.

Main Methods:

• Downloaded 1000 MS data sets from the ProteomeXchange database.
• Utilized pFind software for peptide identification.
• Validated findings using RNC-seq data and compared protein properties with PE1 proteins.

Main Results:

• Identified peptides corresponding to 1788 PE2-PE5 proteins.
• Confirmed 11 PE2 and 16 PE5 proteins with at least two peptides, meeting HPP guidelines.
• Found translation evidence for 16 of these proteins in RNC-seq data, supporting their existence.

Conclusions:

• Reanalyzing large-scale MS data repositories with advanced algorithms is a valuable strategy for identifying PE2-PE5 proteins.
• Multi-dataset peptide identifications can support the existence of PE2-PE5 proteins and warrant further validation.
• High-throughput approaches are crucial for discovering more uncharacterized human proteins.