PeptideReranking

PeptideReranking reorders Peptide-Spectrum Matches (PSMs) to improve identification accuracy in mass spectrometry-based proteomics.

Key Features:

• PSM reranking: Re-ranks Peptide-Spectrum Matches to distinguish true PSMs from false positives.
• PPMRanker: Uses Protein-Peptide Map (PPM) information from the protein database to analyze spatial and structural relationships within proteins.
• PPIRanker: Relies on Precursor Peak Intensity (PPI) data from MS1 spectra to discriminate correct and incorrect identifications based on precursor ion abundance.
• MIRanker: Integrates PPM and PPI data to provide combined evidence for reranking and improve performance.
• Benchmarking: Tested on standard protein mixtures and human/mouse datasets, with PPMRanker and MIRanker reported to outperform PeptideProphet, PeptideProphet+NSP, and SRPI.

Scientific Applications:

• PSM precision enhancement: Improves precision of peptide identifications in mass spectrometry datasets.
• Protein quantification and characterization: Supports more reliable protein quantification and characterization for studying biological processes and disease mechanisms.
• Cross-dataset benchmarking: Applicable to standard protein mixtures and human/mouse proteomic datasets for method comparison.

Methodology:

Reranking of PSMs using three algorithms: PPMRanker (Protein-Peptide Map from the protein database), PPIRanker (Precursor Peak Intensity from MS1 spectra), and MIRanker (integration of PPM and PPI).