Protein and peptide identification algorithms using MS for use in high-throughput, automated pipelines

Area of Science:

• Proteomics and Mass Spectrometry Data Analysis

Background:

• Current proteomics experiments rapidly produce large datasets, outpacing analytical capabilities.
• Existing reviews cover peptide and protein identification methods using mass spectrometry (MS), but lack comparative performance analyses.
• The increasing demand for high-throughput, automated identification systems necessitates efficient and accurate algorithms.

Purpose of the Study:

• To review and compare core algorithms for peptide and protein identification.
• To assess the relative performance of various identification algorithms in terms of accuracy and computational efficiency.
• To address the lack of standardized performance reporting in the literature for these algorithms.

Main Methods:

• Review of core algorithms for peptide mass fingerprinting (PMF), MS/MS database searching, sequence tag searches, and de novo sequencing.
• Comparative assessment of the performance of selected algorithms.
• Proposal for a standardized reporting system for algorithm performance evaluation.

Main Results:

• Identified key algorithms for various peptide and protein identification tasks.
• Assessed the comparative performance of these algorithms, noting variations in accuracy and efficiency.
• Found limited standardized performance data in existing literature.

Conclusions:

• There is a critical need for standardized reporting of peptide and protein identification algorithm performance.
• The adoption of standardized reporting based on freely available datasets is recommended.
• Initial suggestions for the format and content of such datasets are presented.