Identifying Differential Gene Sets Using the Linear Combination of Genes with Maximum AUC

• Activity Number: 515
• Type: Contributed
• Date/Time: Wednesday, August 3, 2011 : 10:30 AM to 12:20 PM
• Sponsor: International Chinese Statistical Association
• Author(s): Chen-An Tsai*+ and Zhanfeng Wang and Yuan-chin Ivan Chang
• Companies: China Medical University and Academia Sinica and Academia Sinica
• Keywords: Gene Set Enrichment Analysis ; discriminatory power ; ROC ; AUC

Abstract

Gene set enrichment analyses (GSEA) provide a useful and powerful approach to identify differentially expressed gene sets with prior biological knowledge. Several GSEA algorithms have been proposed to perform enrichment analyses on groups of genes. However, little attention has been given to the discriminatory power of gene sets. Thus, it is of great interest to identify which differential gene sets are strongly associated with phenotypic class distinction ability by integrating gene expression data with prior biological knowledge. We propose two methods to identify differential gene sets using the area under the receiver operating characteristic (ROC) curve (AUC) of linear risk scores of gene sets, which are obtained through a parsimonious threshold-independent gene selection method within gene sets. The p-values of AUC-based statistics and the AUC values obtained from cross-validation of the linear risk scores are calculated, and used as indexes to identify differential gene sets. The discrimination powers of gene sets are summarized and gene sets that possess discrimination power are selected via a prescribed p-value threshold or a predefined cross-validation AUC threshold.