consensusOrderMEs: Put close eigenvectors next to each other in several sets.

Description

Reorder given (eigen-)vectors such that similar ones (as measured by correlation) are next to each other. This is a multi-set version of orderMEs; the dissimilarity used can be of consensus type (for each pair of eigenvectors the consensus dissimilarity is the maximum of individual set dissimilarities over all sets) or of majority type (for each pair of eigenvectors the consensus dissimilarity is the average of individual set dissimilarities over all sets).

Usage

consensusOrderMEs(MEs, useAbs = FALSE, useSets = NULL, 
                  greyLast = TRUE, 
                  greyName = paste(moduleColor.getMEprefix(), "grey", sep=""), 
                  method = "consensus")

Value

A vector of lists of the same type as MEs containing the re-ordered eigengenes.

Arguments

MEs: Module eigengenes of several sets in a multi-set format (see checkSets). A vector of lists, with each list corresponding to one dataset and the module eigengenes in the component data, that is MEs[[set]]$data[sample, module] is the expression of the eigengene of module module in sample sample in dataset set. The number of samples can be different between the sets, but the modules must be the same.
useAbs: Controls whether vector similarity should be given by absolute value of correlation or plain correlation.
useSets: Allows the user to specify for which sets the eigengene ordering is to be performed.
greyLast: Normally the color grey is reserved for unassigned genes; hence the grey module is not a proper module and it is conventional to put it last. If this is not desired, set the parameter to FALSE.
greyName: Name of the grey module eigengene.
method: A character string giving the method to be used calculating the consensus dissimilarity. Allowed values are (abbreviations of) "consensus" and "majority". The consensus dissimilarity is calculated as the maximum of given set dissimilarities for "consensus" and as the average for "majority".

Author

Peter Langfelder, Peter.Langfelder@gmail.com

Details