WGCNA

WGCNA constructs and analyzes weighted gene co-expression networks from high-dimensional datasets (e.g., microarray and other high-throughput gene expression data) to identify modules of co-expressed genes and relate them to external sample traits.

Key Features:

• Network construction: Constructs weighted correlation networks from microarray and other high-throughput gene expression data.
• Weighted correlations: Uses weighted correlations to quantify pairwise relationships and capture graded co-expression strengths.
• Module detection: Identifies modules (clusters) of highly correlated genes or variables.
• Module summarization: Summarizes modules using module eigengenes (the first principal component) or intramodular hub genes.
• Relating modules to traits: Associates modules with external sample traits using eigengene network methodology.
• Module membership measures: Calculates module membership measures to quantify gene-to-module association strength.
• Utility functions: Provides functions for data manipulation and visualization of networks and modules.
• Interfacing: Supports interfacing with external software for integration into analysis workflows.
• Versatility: Applies its data mining techniques beyond gene expression to diverse high-dimensional biological datasets.

Scientific Applications:

• Cancer research: Identify co-expression modules and candidate biomarkers or therapeutic targets from tumor gene expression data.
• Mouse genetics: Analyze co-expression patterns in mouse genetic studies to link modules to phenotypes.
• Yeast genetics: Characterize modules in yeast genetics datasets to reveal functional gene groups.
• Brain imaging data analysis: Apply co-expression network concepts to brain imaging-derived molecular or gene expression datasets.
• Gene screening and biomarker discovery: Facilitate gene screening methods aimed at discovering candidate biomarkers or therapeutic targets through module detection.

Methodology:

Construction of correlation networks using weighted correlations to capture interdependencies among genes or variables; detection of modules of highly correlated features; summarization of modules via module eigengenes (first principal component) or intramodular hub genes; association of modules with external traits via eigengene network methodology; and calculation of module membership measures.