Can't WGCNA classify the correlation between genes as positive or negative?

Question

[pharamac]

Hello. Nice to meet you. I have one small question about using WGCNA.

A network was created from modules grouped by WGCNA. I know that WGCNA can determine eigengenes with genes belonging to one certain modules and correlate them with traits.

What I'm curious about, not this, is the correlation "between the genes" that make up the module. (For examples, if A, B and C genes build one module I want to display positive or negative correlations between A-B, B-C and A-C... with different colours meaning positive or negative correlations.)

Can't the correlation between the genes that make up the module be displayed as positive or negative? (I planned to use Weight value but Weight is known as a value representing co-occurrence, which has a value between 0 and 1 (only positive) different from this.)

I appreciate your help.

[andres.firrincieli]

you can use the cor() function to recalculate the correlation between the genes of the same module (see this quick tutorial).

Keep in mind that this makes sense only if the network is unsigned since only in this case gene sharing negative and significant correlations are included in the same module.