Comparison at the peptide level with post-translational modification consideration reveals more differences between two unenriched samples

• Published in: Rapid communications in mass spectrometry 2014-06, Vol.28 (12), p.1364-1370
• Main Authors: Yin, Jianrui, Shao, Chen, Jia, Lulu, Gao, Youhe
• Format: Article
• Language: English
• Subjects: Amino Acid Sequence; Biotechnology; Cell Line, Tumor; Humans; Inference; Mass Spectrometry; Molecular Sequence Data; Peptide Mapping; Peptides; Peptides - chemistry; Peptides - metabolism; Protein Processing, Post-Translational; Proteins; Proteins - chemistry; Proteins - metabolism; Proteomics; Searching; Websites
• ISSN: 0951-4198; 1097-0231
• DOI: 10.1002/rcm.6911

RATIONALE In shotgun strategies, peptide sequences are first identified from tandem mass (MS/MS) spectra, and the existence and abundance of the proteins are then inferred from the peptide information. However, the protein inference step can produce errors and a loss of information. To identify the information that is lost using the traditional approaches, this study compared the proteomic data of two leukemia cell lines (Jurkat and K562) at the peptide level with consideration of post‐translational modifications (PTMs). METHODS The raw files from the two cell lines were searched against the decoy IPI‐human database version 3.68, which contains forward and reverse sequences. Then the observed modification name in the results was matched with the modification classification on the Unimod website by a manual search. Only the peptides with ‘post‐translational’ modifications were compared between the two cell lines. RESULTS After searching the database with consideration of PTMs, a total of 44046 non‐redundant peptides were identified in both the Jurkat and K562 cell lines. Of these peptides, even without specific PTM enrichment, 11.43% of them (with at least two spectra in one cell line) existed in different PTM forms between the two cell lines, and 1.73% of the peptides were modified in both cell lines, but with different modifications or possibly on different sites. CONCLUSIONS Comparing proteomic data at the peptide level with consideration of PTMs can reveal more differences between two unenriched samples. Copyright © 2014 John Wiley & Sons, Ltd.