Inferring Gene Functional Modules and Association Networks Using Sparse Canonical Correlation Analysis

• Activity Number: 215
• Type: Topic Contributed
• Date/Time: Monday, August 4, 2014 : 2:00 PM to 3:50 PM
• Sponsor: WNAR
• Author(s): Haiyan Huang*+
• Keywords: sparse canonical correlation analysis (SCCA) ; partial correlation ; community structure ; gene association networks ; gene functional modules

Abstract

Networks pervade many disciplines of science for analyzing complex systems with interacting components. In particular, this concept is commonly used to model interactions between genes and identify closely associated genes forming functional modules. However, characterizing gene relationships remains a challenging problem due to the complexity of high dimensional biological data. In this paper, we develop a two-stage procedure for estimating gene association networks and inferring gene functional groups as tightly connected communities. We propose a novel way of computing edge weights in a gene network, which relies on sparse canonical correlation analysis with repeated subsampling and random partition. Based on the estimated network, community structures can be identied using the block model or hierarchical clustering. Our approach is conceptually appealing as it aims to capture gene relationships in a more realistic manner by taking into account higher level group interactions. It is also flexible enough to incorporate prior biological knowledge. Comparisons with several popular approaches using simulated and real data show our procedure improves both the statistical significanc