A Subspace Module Extraction Technique for Gene Expression Data

Abstract

AbstractConstruction of co-expression network and extraction of network modules have been an appealing area of bioinformatics research. In literature, most existing algorithms of gene co-expression network extract network modules where all samples are considered. In this paper, we propose a method to construct a co-expression network based on mutual information and to extract network modules defined over a subset of samples. The method was applied over several real life gene expression datasets and the results are validated in terms of p value, Q value and topological properties.KeywordsCo-expression networknetwork modulesmutual informationtopological property