In a WGCNA analysis of transcriptome and proteome of a white blood cell in development (in 6 stages), I find in most modules (especially the large ones) over, as well as underexpressed features, but I am using a signed network type.
I attach some of the graphs and results below. This is my first time running a WGCNA analysis and it's hard for me to deduct meaning from some of the graphical outputs. Would it be great if you could give me tips to better understand the outputs and explain why I get a mix of over and underexpressed features? A final question would be if it matters if some of the groups are highly similar. (eg: the last 3 stages are highly similar on proteome level). Does this throw off the analysis? Should I unite them into one group? Would it matter if that combined group is larger than the other groups? (4-5 samples per group, new group would be 14 samples).
Data: transcriptome as tpm with vsn normalisation proteome as intensity with vsn normalisation. Both data batch effect free.
Thank you very much!
(ps: his is a re-post from a question I posted at the bioconductor forum, where it unfortunatly did not get an answer: https://support.bioconductor.org/p/9152480/