Hello! I am using WGCNA for constructing Coexpression Networks . I am now at the point where I am prioritizing genes for candidate hubs. I saw that there are different methods for choosing hub genes in a module. I saw that some people have reported using the 0.8 cutoff for MM ( correlation of the gene with the Module Eigen of the module). But I can't get any hubs for most module that can reach to 0.8. The maximum MM I can see from my modules is around 0.6. What does this suggest, that the modules are not stable enough? Should I just use the intramodular connectivity strength for hubs. Is there a cut of for that value as well?

Thanks!

I'm not sure if a value of 0.6 suggests anything about the stability of module calculation. You could make a histogram of module membership scores and calculate a z-score, for example, for that gene. So if you see that despite a score < 0.8, the MM score of that gene still scores highly as per z-score, AND is also ranked highly by connectivity measures, then it could be a good indication for hubness.