Entering edit mode
5.8 years ago
baldauf
▴
40
Hi, I am still trying to apply the consensus WGCNA to my own data following the Tutorial written by Peter Langfelder and Steve Horvath. I managed to draw the preservation heatmap etc, which corresponds to Figure 1 in the Tutorial II.5 (https://labs.genetics.ucla.edu/horvath/CoexpressionNetwork/Rpackages/WGCNA/Tutorials/Consensus-EigengeneNetworks.pdf). If I understood correctly the consensus modules are the preserved ones between both data sets.
But how do I obtain the differentially connected genes on my two data sets?
I also have a large number of consensus modules (=88), which makes these heatmaps a bit confusing. Do you have here any recommendations?
Thanks for your advices, Jutta
How do you define the differentially connected genes? Genes that apear in the preserved consensus modules in both datasets? Or genes that change their connectivity and direction in each dataset?.
You can merge the clusters (now I don't remember in which tutorial is this explained, but there is one explaining it).
I am not sure how I should define "differentially connected genes". If I understood correctly, than the ones which show up in the preserved consensus modules in both data sets are not differentially connected (because of the word "preserved"). Or am I wrong? So, according to these assumptions of mine, I would determine the genes that are differentially connected which change their connectivity and direction between the two data sets. But I don't know if this is right and how to extract them.
With merging the clusters do you mean Tutorial II.3 ("Relating the consensus modules to female set-specific modules (this section requires the results of Section 2.a of the female turorial"; https://labs.genetics.ucla.edu/horvath/htdocs/CoexpressionNetwork/Rpackages/WGCNA/Tutorials/Consensus-RelateToFemMods.pdf )?
You could then use DiffCoExp or DiffCoEx (not the same tools) that depend on WGCNA.
No, I was referring to the 2.b.5 section.
ok, thanks! I will have a look at these tools and try ;)