Comparing PeptideShaker Plugins and Features: Which Ones Matter?

PeptideShaker — A Beginner’s Guide to Proteomics Analysis

What it is

PeptideShaker is an open-source Java application for interpreting, validating, visualizing, and re‑analyzing peptide and protein identification results from MS/MS search engines (e.g., X!Tandem, MS-GF+, Comet, Mascot, Andromeda). It aggregates search-engine outputs, performs validation (FDR filtering, protein inference), and produces interactive reports and exportable files (mzIdentML, TSV, plots).

Key features

• Multi-engine support: imports results from many search engines and mzIdentML.
• Validation & FDR control: peptide and protein level filtering and confidence scoring.
• Protein inference & grouping: resolves peptides-to-proteins relationships.
• PTM inspection: view and localize post‑translational modifications.
• Visualization: spectrum viewer, peptide/protein reports, QC plots.
• Annotation: GO-term enrichment and mapping to structures (PDB) where available.
• Export: detailed reports, mzIdentML, and project packages for the desktop viewer.
• Command-line & GUI: desktop application with CLI options; Java-based, cross-platform.
• License: Apache 2.0 (free).

Typical workflow (prescriptive)

1. Convert raw LC-MS/MS files to MGF or mzML (e.g., ProteoWizard msConvert).
2. Run one or more search engines (via SearchGUI or standalone) using a FASTA database (include decoys for FDR).
3. Import search results and the FASTA into PeptideShaker.
4. Set FDR thresholds and validation parameters; run the validation.
5. Inspect peptide-spectrum matches (PSMs), PTMs, and protein groups in the viewer.
6. Generate reports (TSV, mzIdentML) and QC plots; export a PeptideShaker project zip for sharing.

System & resources

• Java required (desktop: macOS, Linux, Windows). Web/online interfaces (PeptideShaker Online) exist for cloud use.
• Documentation and downloads: CompOmics / PeptideShaker project pages and tutorials (e.g., Galaxy training).

Practical tips

• Always include decoy sequences and set appropriate FDR (commonly 1% at peptide/protein level).
• Use SearchGUI to run multiple search engines for improved coverage.
• Inspect single-peptide protein identifications carefully — consider requiring ≥2 peptides for confident protein calls.
• For large datasets, increase Java memory (e.g., -Xmx) when launching the app.

Further reading / resources

• PeptideShaker publications (Vaudel et al., Nat Biotech 2015) and online documentation/tutorials (CompOmics, Galaxy training).