Consensus wgcna is aimed to generate common networks across different data sets. When we identified consensus modules, how to clarify the relationships between these consensus modules. One method I've known is to calculate eigengenes and their correlations in each data set, and then for each data set, a eigengene network is formed. But I want to generate a common correlation network for all the data sets, not different eigengene network for different data set. 

One of my solution is to choose the minimum of eigengene correlation across all the data sets (as my consensu TOM is the minimum of TOM across all data sets). Is their any other method to depict a common consensus module relationship?

And a further question, how to depict the significance of the relationship?