WGCNA Analysis - blockwiseModules data processing

Question

[Raito92]

Good evening, I am currently running a WGCNA analysis. As suggested by a colleague, I switched from regular single-block WGCNA calculation to blockwiseModules, due to large (42,000 genes) dataset size. By setting the max block dimension to 40,000 , I obtain two separate blocks. Also, TOM matrices are exported and saved separately.

I don’t know how to proceed at this point. The function produced some data objects, but also exported two separated TOMs matrices.

What should I do? Most of the downstream functions explicitly require to work on a TOM (exporting data to Cytoscape for instance). I can easily import the matrix back into R, but I have two of them (same as number of blocks). Do I need to merge them somehow? Or should work on the two matrices separately? If so, how is this done? The tutorial doesn’t really expand on this.

Thank you in advance!

[cfos4698]

Some of the discussion I had previously re: exporting/saving/combining TOMs here might help: https://support.bioconductor.org/p/125161/

[OrtegaC]

Hello, Maybe it's too late for this answer, sorry...

I was looking for Horvath's website, but it appears unavailable (I don't know if it was mired or another general problem).

https://horvath.genetics.ucla.edu/html/CoexpressionNetwork/Rpackages/WGCNA/Tutorials/Consensus-NetworkConstruction-blockwise.pdf

This was my code:

# Block-wise network

cor <- WGCNA::cor
bwnet = blockwiseModules(datExpr, power = 24, 
TOMType = "signed", 
maxBlockSize = 5000,  
minModuleSize = 30,
reassignThreshold = 0,
mergeCutHeight = 0.20, 
numericLabels = TRUE, pamRespectsDendro = FALSE,
saveTOMs = F, 
verbose = 3)

save(bwnet, file = "1_AtTOM-blockwise.RData")
#load("1_AtTOM-blockwise.RData")
cor<- stats::cor
save.image("part2_TOM.RData") #save all objects

#str(bwnet)
summary(bwnet)
# get number of genes for each module
table(bwnet$colors)

###Plot module_tree_blockwise
# open a graphics window
#sizeGrWindow(12, 9)

mergedColors = labels2colors(bwnet$colors)

# Plot the dendrogram and the module colors underneath
pdf(file = "4_Plot_module_tree_blockwise.pdf", width = 8, height = 6);
plotDendroAndColors(bwnet$dendrograms[[1]], mergedColors[bwnet$blockGenes[[1]]],
                    "Module colors",
                    dendroLabels = FALSE, hang = 0.03,
                    addGuide = TRUE, guideHang = 0.05)
dev.off()

#
moduleLabels = bwnet$colors
moduleColors = labels2colors(bwnet$colors)
MEs = bwnet$MEs;
geneTree = bwnet$dendrograms[[1]];
save(MEs, moduleLabels, moduleColors, geneTree, 
     file = "Consensus-NetworkConstruction-auto_AtTOM.RData")
#This file saved can be used to export the nwt to cytoscape
###https://horvath.genetics.ucla.edu/html/CoexpressionNetwork/Rpackages/WGCNA/Tutorials/FemaleLiver-06-ExportNetwork.pdf

[rpolicastro]

The internet archive has a cached version

https://web.archive.org/web/20230203001746/https://horvath.genetics.ucla.edu/html/CoexpressionNetwork/Rpackages/WGCNA/Tutorials/Consensus-NetworkConstruction-blockwise.pdf