Functional module extraction by ensembling the ensembles of selective module detectors

A group of functionally related genes constitutes a functional module taking part in similar biological activities. Such modules can be employed for the interpretation of biological and cellular processes or their involvement in associated diseases. Detection of such modules from gene expression data is a difficult task, but important from system biology point of view. Different module detectors have been proposed for a few decades with their relative merits and demerits. They can be broadly classified as Clustering, Bi-Clustering and Network based. In this work, we try to combine the merits of some of the selective module detectors picked from three types of module detectors. We perform a two-level ensemble by unifying the goodness of different module detectors. For our experimentation, we use RNAseq read counts as a measure of gene expression. We compare ensemble outcomes with state-of-the-art module detectors and observe a superior performance in comparison to them.