Clustering Procedures and Module Detection

Abstract

This chapter describes a case study and R code for carrying out an integrated weighted correlation network analysis of mouse gene expression, sample trait, and genetic marker data. It describes how to (a) use sample networks (signed correlation networks) for detecting outlying observations, (b) find co-expression modules and key genes related to mouse body weight and other physiologic traits in female mice, (c) study module preservation between female and male mice, (d) carry out a systems genetic analysis with the network edge orienting approach to find causal genes for body weight, and (e) define consensus modules between female and male mice. We also describe methods and software for visualizing networks and for carrying out gene ontology enrichment analysis.