UniScore, a Unified and Universal Measure for Peptide Identification by Multiple Search Engines.

Area of Science:

• Proteomics
• Mass Spectrometry
• Bioinformatics

Background:

• Data-dependent acquisition (DDA) in LC/MS/MS proteomics generates complex datasets.
• Standardizing and integrating outputs from multiple search engines remains a challenge.
• Existing rescoring methods may require substantial computational resources.

Purpose of the Study:

• To introduce UniScore, a novel metric for standardizing and integrating search engine outputs in DDA proteomics.
• To evaluate UniScore's performance against conventional methods.

Main Methods:

• UniScore calculates peptide matches using only product ion annotation information.
• It employs a false discovery rate (FDR) based on the target-decoy approach for acceptance criteria.
• The method was tested on large-scale global proteome and phosphoproteome datasets.

Main Results:

• UniScore demonstrated superior performance compared to individual search engines (Comet, X! Tandem, Mascot, MaxQuant).
• It processed large datasets efficiently with minimal computational resources.
• UniScore successfully performed peptide matching in chimeric spectra without additional filters.

Conclusions:

• UniScore offers an effective and efficient approach for integrating and standardizing proteomics search engine results.
• The metric is robust, scalable, and applicable to complex proteomic datasets, including chimeric spectra.