Abstract
The transcriptional regulation of cellular functions is carried out by the overlapping functional modules of a complex network. In this paper, a statistical approach for detecting functional modules in the transcriptional regulatory networks (TRNs) is studied. The proposed method defines modules as groups of links rather than nodes since nodes naturally belong to more than one module. Furthermore, the proposed algorithm is evaluated on the Escherichia coli TRN. The experimental results demonstrate that it detected a suitable number of overlapping modules that were biologically meaningful without any prior knowledge about the modules.