Entering edit mode
4 months ago
eduardop.dev
•
0
Hello everyone,
I am interested in applying co-expression network analysis to bulk rnaseq data downloaded from ncbi. The problem is I am quite unsure about the structure of the expression matrix used as input for the WGCNA package. I know that the rows correspond to genes and the columns are samples/traits. But what kind of samples can I (or should I) put in the same matrix? Does it make sense for example for all the columns to correspond to the same species under the same type of stress? Should I also put samples where the species is not under any type of stress?
The documentation goes over this but generally if there is a large effect size between the two conditions, and you have 15+ samples in each condition, you should process them separately at first. Later on in the analysis you can start doing comparisons between the networks.
WGCNA tutorial is available in a new location: https://bioinformatics.stackexchange.com/questions/21885/where-to-access-the-wgcna-tutorial-documents-horvath-lab-site-down/21886#21886