MultiGlobAl: Global alignment of multilayer networks

Abstract

In bioinformatics, networks are used for modeling biological objects (e.g., genes, proteins, drugs) as nodes, and their interactions as edges. A multilayer network is used for modeling multiple semantic levels. It consists of nodes and edges distributed over multiple layers. Network Alignment (NA) is a methodology that allows mapping nodes between two or more networks of interest, by preserving topologically similar regions; for instance, it is used in biology to transfer knowledge from one biological species to another. Traditional NA methodologies are ineffective for multilayer networks, in which these are unable to recognize interlayer edges, as well as to inspect the whole topology. We present Multilayer Global-Aligner (MultiGlobAl), a software tool for the Global NA (GNA) of multilayer networks. It features nodes based on their interactions (both interlayer and intralayer edges), in respect to the whole topology of the network. This information is used for building the similarity matrix, on which the NA function is applied. Our solution is able to perform the one-to-one node mapping between two multilayer networks of interest, by maximizing the objective function between all pairs of nodes.