moduleMemberCorrelations: moduleMemberCorrelations

In Gibbsdavidl/ProCoNA: Protein co-expression network analysis (ProCoNA).

Description

Computes the relation between peptides and eigenvector summaries and also peptides and phenotypes.

Usage

moduleMemberCorrelations(pnet, pepdat, phenotypes)

Arguments

pnet	The peptide net object
pepdat	The peptide data matrix
phenotypes	The matrix of traits

Value

Matrix of Pearson correlations with peptides in rows.

Author(s)

David L Gibbs

Examples

data(ProCoNA_Data)
#net1 <- buildProconaNetwork("peptide network", peptideData)
n <- length(samples(net1))
phenotypes <- matrix(rnorm(10*n), nrow=60)
pepcor <- moduleMemberCorrelations(net1, peptideData, phenotypes)

# To plot the heatmap: 
# moduleCors <- correlationWithPhenotypesHeatMap(net1, phenotypes, modules = 1:5, 
#    plot = NULL, title = "Module-trait relationships", textSize = 0.5)

#########################################################################
# quick function to write out the tables for specific modules.
#moduleData <- function(pepnet, pepcors, module, pepinfo, fileprefix) {
#  moduleX <- pepnet@peptides[which(pepnet@mergedColors==module)]
#  moduleInfo <- pepinfo[which(pepinfo$Mass_Tag_ID %in% moduleX),]
#  moduleCors <- pepcors[which(pepcors$Module==module),]
#  corname <- paste(fileprefix, "_correlations.csv", sep="")
#  write.table(moduleCors, file=corname, sep=",", row.names=F)
#  infoname <- paste(fileprefix, "_peptide_info.csv", sep="")
#  write.table(moduleInfo, file=infoname, sep=",", row.names=F)
#}
########################################################################

# WRITE OUT A TABLE WITH THE BELOW FUNCTION CALL  :)#
# moduleData(peptideNetwork, pepcor, 1, masstagdb, "Module_1")

Gibbsdavidl/ProCoNA documentation built on May 8, 2019, 7:51 p.m.