|
Abstract:
|
Network analysis has become a very active area of research for analyzing complex data. Many networks including biological networks are bipartite, where nodes are divided into two sets, and only connections between nodes of different sets are allowed. Examples of bipartite biological networks include gene-disease network, transcription factors and binding sites, host-pathogen interactions and plant-pollinator.
Motivated by the challenges and the lack of model-based approaches in detecting communities in a bipartite network with node covariates, we propose a new statistical method for matched community detection in a bipartite network with node covariates. We show the efficiency of our algorithm through synthetic and real data, and we also discuss the wide applicability of our method to studying gene orthology networks with gene expression as covariates.
|
