WGCNA - Using real data for tutorial III
2
0
Entering edit mode
@hywelwilliams-9516
Last seen 8.2 years ago

I have worked through the WGCNA tutorial I using my own data and I have used method 2b to construct my network and form modules and i have progressed to stage 3 where I've applied my traits and I have found some interesting findings.

I now want to look at hub genes from within those modules significantly associated with my trait of interest but when I move to the Tutorial III section I find that the input files bear no relationship to the files saved from the Tutorial I. Specifically, I am struggling to find a match from my data that corresponds to ModuleEigengeneNetwork1, GeneSummaryTutorial.csv, and TraitsTutorial.csv.

 

My aim is to get to Tutorial III part 7 and any guidance as to how I can get the data from Tutorial I to match the required input for Tutorial III would be greatly appreciated.

 

Many thanks - Hywel

pathways wgcna network rna-seq • 2.7k views
ADD COMMENT
0
Entering edit mode
@peter-langfelder-4469
Last seen 9 months ago
United States

I am a bit confused by your description. Are you working through tutorial I? If so, you should be looking at section 3, not Tutorial III, for an example of how to find important genes.

If you want to use Tutorial III, the best thing to do is to adapt the code in Section 2 ("Loading of expression data") so it loads and properly formats your own data.

For moduleEigengeneNetwork, use cbind() on the trait of interest and the module eigengenes (part of the output of blockwiseModules or you can use moduleEigengenes(datExpr, moduleColors) if you need to calculate them).

For the actual datExpr, you can use what you have from Tutorial I.

For traits, use a file containing your sample traits (conditions). The tutorial uses the variable 'y' to hold the trait of interest.

The tutorial is confusing (my fault) regarding GeneSummaryTutorial. This file is actually generated in Section 7, so we should not have been loading it in Section 2. The only thing it contains that's needed in the following is "truemodule" for which you can substitute your module colors (or labels). Just be aware that this is not really the 'true' module since in real data we don't really know 'true' modules.

HTH,

Peter

ADD COMMENT
0
Entering edit mode
@hywelwilliams-9516
Last seen 8.2 years ago

Dear Peter, that is very helpful, I will use your suggestions and repeat the analysis. Best wishes - Hywel

ADD COMMENT

Login before adding your answer.

Traffic: 614 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

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

Powered by the version 2.3.6