Dear Professor Peter Langfelder and all members,

I'm using WGCNA to indentify co-expressed genes from my RNAseq data. I first started with automatic: one-step network construction, but ended up having too many modules and I found in step-by-step, there is a nice tutorial and code chunks on how to use dynamic cut tree and merge module with similarity and I ended up having less modules (around 18 modules) compared with 40 modules using automatic approach.

I have watched WGCNA tutorial published in youtube and professor suggested to use singed network as this tends to generate more biological meaning modules compared with unsinged (correct me if im wrong.

With the step-by-step approach, I could not find where I could incorporate singed or unsigned network in the tutorial at all. I am wondering whether professor or members could assist me with it as I am very keen to test the singed network for my data in step-by-step approach because I do not get any interesting output from unsigned network.

https://horvath.genetics.ucla.edu/html/CoexpressionNetwork/Rpackages/WGCNA/Tutorials/FemaleLiver-02-networkConstr-man.pdf

Can I also perform dynamtic cuttree and merge modules with similarity in automatic approach? thank for all your help in advance..

Kind Regards,

synat