FineFDR

FineFDR implements taxonomy-specific false discovery rate (FDR) control for metaproteomics by grouping peptide-spectrum matches, peptides, and proteins by taxonomic origin and computing FDR within each group to reduce false positives in MS/MS-based protein identification.

Key Features:

• Taxonomy-specific FDR control: Groups identified entities by taxonomic origin and computes FDR separately within each taxonomic group.
• Multilevel assessment: Applies FDR control at the peptide-spectrum match (PSM), peptide, and protein levels.
• MS/MS database-search compatibility: Operates on tandem mass spectrometry (MS/MS) spectra matched against protein sequence databases for peptide and protein annotation.
• Extension of target-decoy approach: Incorporates taxonomic grouping into target-decoy FDR estimation rather than treating all identifications uniformly.
• Empirical performance: Demonstrates higher precision in peptide and protein identification compared to Comet, Percolator, TIDD, and Tailor on simulated and real-world datasets.

Scientific Applications:

• Metaproteomic protein identification: Improves accuracy of peptide and protein identification in metaproteomic analyses of microbial communities using MS/MS.
• Protein quantification support: Reduces false positives in identification steps that underpin protein quantification in complex microbial samples.
• Method benchmarking: Provides a taxonomy-aware framework for comparing search and validation methods such as Comet, Percolator, TIDD, and Tailor.

Methodology:

Matches MS/MS spectra against protein sequence databases, uses the target-decoy approach for FDR estimation, stratifies identifications by taxonomic origin and computes FDR separately within each taxonomic group, and evaluates performance on simulated and real-world datasets versus Comet, Percolator, TIDD, and Tailor.