Crux

Crux analyzes protein tandem mass spectrometry (MS/MS) datasets to identify and statistically validate peptide and protein matches for proteomics studies.

Key Features:

• Database search program: Extends the Sequest algorithm and uses a peptide indexing scheme for rapid retrieval of candidate peptides matching spectra.
• Decoy peptide generation: Generates shuffled decoy peptides dynamically from each target database entry to provide a null model for false discovery rate (FDR) estimation.
• Weibull distribution-based p value calculation: Fits a Weibull distribution to observed scores to compute p values for peptide-spectrum matches.
• Semisupervised discrimination method: Applies a semisupervised learning method to distinguish target from decoy matches and improve identification rates.
• Performance optimization: Employs efficient computational strategies to improve speed and accuracy for large-scale proteomic datasets.
• Implementation: Implemented in C.

Scientific Applications:

• Protein identification and quantification: Identifies and quantifies proteins from complex biological samples using MS/MS data.
• Large-scale proteomics processing: Processes large volumes of MS/MS data for proteome-wide studies.
• Biological discovery: Supports disease biomarker discovery, protein interaction network analysis, and functional genomics studies.

Methodology:

Extends a Sequest-like database search with a peptide indexing scheme, generates shuffled decoy peptides per target entry, fits Weibull distributions to score distributions to calculate p values, and applies a semisupervised discrimination method to separate target and decoy matches; implemented in C.