Since I've just started my analysis using WGCNA, I'm not really familiar with it so far. I need to do different comparisons in WGCNA. I have two ethnicities, for each of them I have three comparisons group, let's say A, B , C and among each of these, three classes (class1, class2, class3). For now, I run WGCNA for one ethnicity (all groups) - 8000 genes, after that I did gene set enrichment for each modules and I split my data set by groups while doing heatmaps for particular module. My other approach was to do WGCNA for a particular group (8000 genes). What I was thinking, if this is possible to take a list of 8000 most variable genes for each group A, B, C, then remove the genes that are the same for each group and run WGCNA on all of it. I'm using R, but I'm not sure how I should do that and which of these methods makes most sense.
Also, my question is, should I run DESeq2 on my data set before doing WGCNA? Thanks for help
1 answer
One may use the output of DESeq2 as input to WGCNA. For example, DESeq2 will take your RNA-seq raw counts as input, normalise them, and then transform them to regularised log and/or variance stabilised counts. Using the normalised counts, DESeq2 will also perform differential expression comparisons between your groups of interest. I will assume that you know how to do all of this.
With WGCNA, you should start with the regularised log or variance stabilised counts. You can run it on your entire dataset in an unbiased fashion to derive the modules (with justification, you can run it separately based on ethnicity, too). You can then do enrichment analysis on the genes assigned to each module, as you have already done. You can also correlate the module eigenvalues to your traits to see which eigenvalues statistically significantly correlate with these. I explain this somewhat here: WGCNA modules and categorical traits relationship
Keep in the back of your mind that WGCNA is fundamentally based on correlation. Not everyone openly accepts the results of WGCNA, and also other network analysis tools.
Kevin
Log in to answer this question.
-
DESeq2 multiple group comparisons
written by Bhumi •
-
Using Statistical enrichment test for groups Principal Component Analysis
written by Gabriel
-
DEG analysis grouping help
written by sir.outman •
-
edgeR Section 3.2.6 ANOVA example
written by maynard •
-
Find most upregulated genes in one library?
written by williams.damon26 •
-
Using a chi-squared test for independence for DEG analysis
written by SylviaR •
-
DEseq three-group comparison
written by duvaller •
-
Assigning score/values to a group based on its elements
written by muralinmars •
-
t test PROBLEM
written by yasjas •
-
Pheatmap Annotation
written by James •
Thank you very much for your help, that's exactly what I needed!