The initial assignment of genes to modules using WGCNA is binary: in a given network, each gene will be assigned to one module or zero modules. In vivo, however, a gene may be expressed in multiple cell types or participate in multiple functional pathways. Therefore, to provide a more complete representation of gene coexpression relationships in each network, we summarized each module with a single representative expression profile, which we refer to as the module eigengene (Methods and Supplementary Fig. 2). We then calculated the Pearson correlation between the expression level of each gene on the microarray and each module eigengene. This quantity, which we refer to as a gene’s ‘module membership’, describes the extent to which a gene conforms to the characteristic expression pattern of a module29 (initial module assignments (Fig. 1), complete module membership values and associated P values are reported for all genes in each network in Supplementary Tables 3-6 online).
