Module membership vs gene significance does not correlate (WGCNA)
Entering edit mode
neber • 0
Last seen 3.4 years ago


I am using WGCNA to identify the modules with co-expressed genes and assign these modules to traits.

When I generate a labeled heatmap (please see the code below) to check the module-trait correlation, I see some modules correlating with some traits. However, when I check the module membership vs. gene significance for the correlating modules and traits from the heatmap, there is no correlation (please see an example of the code I used below).

Could someone please explain the reason for getting different results? How should I interpret this result?


The code I used for the labeled heatmap:

labeledHeatmap(Matrix = moduleTraitCor,
               #xLabels = names(datTraits),
               xLabels = cond,
               #yLabels = names(MEs),
               yLabels =namey,
               #ySymbols = names(MEs),
               ySymbols = namey,
               colorLabels = FALSE,
               colors = blueWhiteRed(50),
               textMatrix = textMatrix, 
               setStdMargins = FALSE,
               cex.text = 0.9,
               zlim = c(-1,1))

The code I used for the module membership (MM) vs. gene significance (GS) plot creation:

verboseScatterplot(abs(geneModuleMembership[moduleGenes, column]),
                   abs(geneTraitSignificance[moduleGenes, 1]),
                   xlab = paste("Module Membership in", module, "module"),
                   ylab = "Gene significance for fat",
                   main = paste("Module membership vs. gene significance\n"),
                   cex.main = 1.2, cex.lab = 1.2, cex.axis = 1.2, col = module)
wgcna module membership gene significance • 1.3k views
Entering edit mode
Last seen 6 months ago
United States

First, make sure that it's not a trivial mistake of plotting a different module or trait than the combination that has a high correlation. If that has been checked, lack of GS-MM correlation could indicate that only a submodule relates to the trait. In that case, perhaps redoing the analysis with a finer module splitting could help. Overall, no GS-MM correlation to me suggests to consider the association more tentative, needing further validation or evidence.


Login before adding your answer.

Traffic: 549 users visited in the last hour
Help About
Access RSS

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6