Ranking of network elements based on functional substructures.

Centrality analysis has been shown to be a valuable method for the structural analysis of biological networks. It is used to identify key elements within networks and to rank network elements such that experiments can be tailored to interesting candidates. Several centrality measures have been studied, in particular for gene regulatory, metabolic and protein interaction networks. However, these centralities have been developed in other fields of science and are not adapted to biological networks. In particular, they ignore functional building blocks within biological networks and therefore do not consider specific network substructures of interest. We incorporate functional substructures (motifs) into network centrality analysis and present a new approach to rank vertices of networks. A method for motif-based centrality analysis is presented and two extensions are discussed which broaden the idea of motif-based centrality to specific functions of particular motif elements, and to the consideration of classes of related motifs. The presented method is applied to the gene regulatory network of Escherichia coli, where it yields interesting results about key regulators.