I have a set of biologically relevant transcriptional regulatory networks which are bipartite with a set of transcription factors and the genes that they regulate. However, when I put my network on Cytoscape to get the centrality measures, average path length etc and treat the network as directed, I get a value of either 1 or zero for each node. When I treat the network as un-directed, I get a range of different values for centrality measures etc. I need to compare this network with another one which I have. Does it make sense that I treat the network as un-directed to get the values for comparison because with just 1 and 0 I cant get much. Could you refer to any published papers in which networks have been treated as un-directed for a comparison of topological features.