modulesFromGeneTree: Modules from Gene Tree

In Sage-Bionetworks/SageBionetworksCoex: Gene Coexpression

Description

Computes discrete groups ("modules") of genes, based on a gene dendrogram.

Usage

modulesFromGeneTree(geneTree, expressionData, dichotCor = NULL, tomDist = NULL, dynamicCutMethod = "tree", mergeModuleMethod = "conn")

Arguments

geneTree	The gene dendrogram, returned by clusterGenes()
expressionData	The input data for the analysis, a normalized matrix of gene expression data, given as a data frame with rows as samples and columns as genes/probes.
dichotCor	|cc|^beta, returned by clusterGenes(), only required if mergeModuleMethod=="conn"
tomDist	Topological Overlap Matrix (TOM) expressed as a distance matrix, returned by clusterGenes(), only required if dynamicCutMethod=="hybrid"
dynamicCutMethod	"tree" or "hybrid". The former is the 'classic' approach used at Sage Bionetworks, while the latter is the default for the UCLA-WGCNA approach. The details are here: http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/BranchCutting/Supplement.pdf
mergeModuleMethod	"conn" or "eigen". The former is the 'classic' approach used at Sage Bionetworks, in which a module's representative gene is the most highly CONNected one. The latter is the default for the UCLA-WGCNA approach, in which the representative gene is the module's EIGENvector.

Details

There are two steps to module finding: (1) Cut the dendrogram branches to produce subtrees, the leaves of which become the modules, (2) Rejoining modules which are too close to be considered separate. Each steps can be done by different algorithms. For the "classic" version use dynamicCutMethod="tree" and mergeModuleMethod="conn". For the default method of UCLA-WGCNA use dynamicCutMethod="hybrid" and mergeModuleMethod="eigen". The references for the algorithms are given below.

Value

geneModules	A vector of module memberships, i.e. geneModules[i] is the module to which gene i belongs. "grey" is a reserved name for genes which belong to no module.
genePCTree	the dendrogram of representative vectors of each module

Author(s)

Bruce Hoff

References

http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/

Zhang, B. & Horvath, S. (2005) Statistical Applications in Genetics and Molecular Biology 4, Article 17.

Langfelder P, Zhang B, Horvath S (2007) Defining clusters from a hierarchical cluster tree: the Dynamic Tree Cut library for R. Bioinformatics 2008 24(5):719-720

http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/Rpackages/WGCNA/

Sage-Bionetworks/SageBionetworksCoex documentation built on May 9, 2019, 12:11 p.m.