GenePattern

GenePattern provides computational analysis of diverse biological datasets to enable reproducible investigation of gene expression (RNA-seq and microarray), sequence variation, copy number variations, proteomics (LC-MS), flow cytometry, and network analysis.

Key Features:

• Multi-omics support: Supports analysis of gene expression (RNA-seq and microarray), sequence variation, copy number variations, proteomics, flow cytometry, and network analysis.
• Analytical methods: Integrates hundreds of analytical methods for statistical analysis, quantification, and pattern recognition across data types.
• "Landmark Matching": Implements "Landmark Matching", a time base-independent method that propagates peptide identities onto LC-MS features using historical data from multiple acquisition strategies.
• "Peak Matching": Implements "Peak Matching", which clusters identical molecular species across multiple LC-MS experiments in an identity-independent manner.
• PEPPeR platform: Hosts the Platform for Experimental Proteomic Pattern Recognition (PEPPeR), which extends statistical tools originally used for microarray analysis to proteomics data.
• Reproducibility: Addresses reproducibility challenges arising from chromatographic separation in quantitative proteomics.
• Calibration and quantification: Enables calibration across a wide dynamic range (2.5 orders of magnitude) and precise quantification of ratios in simple and complex mixtures.
• De novo marker discovery and MS/MS identification: Performs de novo marker discovery by assessing statistical significance of unidentified accurate mass components and supports their identification via MS/MS acquisition.

Scientific Applications:

• Biomarker discovery: Facilitates discovery of protein and peptide biomarkers through statistical significance testing of accurate mass components and MS/MS validation.
• Quantitative proteomics: Supports accurate ratio quantification and calibration across mixtures with wide dynamic range.
• Cross-run peptide propagation: Propagates peptide identities across LC-MS experiments using time base-independent methods to improve cross-experiment comparability.
• Omics integration: Enables combined analysis of gene expression, sequence variation, and copy number variation alongside proteomic and flow cytometry data.
• Network and systems analysis: Applies network-analysis methods to elucidate relationships within complex biological systems.
• Marker validation: Associates discovered markers with known proteins whose concentrations were altered, using targeted MS/MS acquisition for identification.

Methodology:

Integrates "Landmark Matching" (time base-independent peptide identity propagation using historical acquisition data) and "Peak Matching" (identity-independent clustering of identical molecular species across LC-MS experiments); implements PEPPeR to extend microarray statistical tools to proteomics; performs calibration across 2.5 orders of magnitude, statistical testing of unidentified accurate mass components for de novo marker discovery, and MS/MS acquisition for marker identification.