Confusion about WGCNA results

Question

[younglin113]

Lately, I have tried to perform WGCNA analysis in R with WGCNA package against a RNA-seq dataset. And in this process, I found the numbers of color modules displayed by labeledHeatmap and Epigegene Network were different. And the code that produced those two plots are:

labeledHeatmap(Matrix = moduleTraitCor,
           xLabels = colnames(design),
           yLabels = names(MEs),
           ySymbols = names(MEs),
           colorLabels = FALSE,
           colors = greenWhiteRed(50),
           textMatrix = textMatrix,
           setStdMargins = FALSE,
           cex.text = 0.5,
           zlim = c(-1,1),
           main = paste("Module-trait relationships"))


plotEigengeneNetworks(MET, "",
                      marDendro = c(0,4,1,2),
                      marHeatmap = c(3,4,1,2),
                      cex.lab = 0.8, xLabelsAngle = 90)

The plots I got are as following:

As you can see, there are 32 modules in labeledHeatmap and only 31 in Epigengene Network. So, what's the secret here or did I miss something? Thanks for your time!!

[younglin113]

Answer 1

[andres.firrincieli]

By default the function plotEigengeneNetworks exclude the grey module which include non-module genes

[younglin113]

Thanks a lot for your response. It helps a lot. But according to the labeledHeatmap, the "grey" module correlated to the AD trait most closely, should I ignore this and focus on "yellow" module for further analysis?

[andres.firrincieli]

See my answer here

Since the clustering algorithm is not perfect, my guess is that the hub of the grey module could actually belong to the yellow module. How many genes do you have in the grey module? What is the output of 1-cor(MEs)?

[younglin113]

There are 1564 genes in the "yellow" module. And the 1-cor(MEs) result between MEgrey and MEyellow is 0.2207696.

[andres.firrincieli]

the dissimilarity is quite low. My suggestion is to merge the modules whose expression profile is very similar (2.b.5)

[younglin113]

I got that. Thanks a lot for your suggestion.