hello, I am playing with network reconstruction, in particular I’m working with coexpression networks in multiple tissues or organisms.
Say I have two matrices for expression data in two different tissues, I surely can infer a network for each tissue or a “global” network joining the two data matrices. I wonder if it is possible to use the whole matrix specifying the tissue identity of each column, as a covariate (if it makes any sense).
You can add the tissue identity as an additional “node” in the reconstruction, but I’m not sure the results will work as you expect.
Adding covariates is something which is in my TODO list.
I’m not sure how the additional node hack would work either.
I think I will continue comparing separate networks until you reach that entry on your TODO list 
Thanks!