Hi,

I have a single network that I have created using WGCNA, is it possible to test module preservation on just the one network. I tried using the following:

#FOR MODULE PRESERVATION

data=read.csv("preservation.csv", header=TRUE) # expression data file, with columns = genes and rows = samples

moduledata=read.csv("modulelist.csv", header=TRUE) # module file, with two columns, one for geneID and one for assigned module
colorsmodule = moduledata$Module

multiExpr=data #this is the expression data
colorList = colorsmodule # this is the list of modules

system.time( {
  mp = modulePreservation(multiExpr, colorList, referenceNetworks=1,
                          nPermutations = 1000,
                          networkType = "signed",
                          randomSeed = 2905,
                          quickCor=0,
                          verbose = 4, 
                          indent = 0)

but it gave the error:

Error in multiExpr[[set]]$data : $ operator is invalid for atomic vectors
Timing stopped at: 2.15 0 2.17 

Which I assume is because it is looking for two sets to compare? Should I just subset my data to say just 5 samples and input this? Or is it just not possible to test for robustness in this case?

Best wishes,

Rebekah

Create a multi-data structure with two copies of your data, perhaps like this:

multiExpr = multiData(Set1 = data, Set2 = data)
colorList = list(Set1 = colorsmodule)

Then run modulePreservation with this input and look at module quality scores in the output (ignore the preservation metrics).

Peter