decoy.pl

decoy.pl generates composite target-decoy protein databases to enable estimation of false discovery rates and improve peptide identification confidence in LC-MS/MS proteomics.

Key Features:

• Decoy Database Generation: Creates composite target-decoy databases by incorporating decoy sequences into protein sequence databases to enable statistical evaluation of peptide-spectrum matches.
• Scoring Criteria Optimization: Supports setting score thresholds to balance sensitivity and specificity, including selection to achieve an estimated 1% false positive rate at maximum sensitivity.
• Compatibility with Spectral Interpretation Software: Produces databases compatible with Mascot and SEQUEST spectral interpretation algorithms.
• Instrument Support: Supports data from linear ion trap (LTQ) and hybrid quadrupole time-of-flight (QqTOF; QSTAR) mass spectrometers.

Scientific Applications:

• FDR estimation and confidence assignment: Estimation of false discovery rates (FDR) and assignment of confidence for peptide and protein identifications in LC-MS/MS datasets.
• Cross-platform comparison: Comparing proteomics results across mass spectrometry platforms such as LTQ and QSTAR.
• Workflow evaluation and reproducibility assessment: Evaluating and comparing proteomics workflows, including yeast proteome measurements, and exploiting variability in replicate data acquisitions to enhance identification confidence.

Methodology:

Generates composite target-decoy databases by integrating decoy sequences into protein databases and applies the target-decoy approach to estimate false discovery rates and set score thresholds (e.g., to an estimated 1% false positive rate) for peptide identification; outputs are compatible with Mascot and SEQUEST and support LTQ and QqTOF (QSTAR) datasets.