Abstract 2000: Biological function annotation of microarray data using correlated gene indices

Abstract

Abstract Although whole-genome microarray analysis has become a routine tool in biomedical research, extracting meaningful information remains a challenge. We describe a method for assessing biological information in microarray data that relies on the observation that genes common to a biological function often exhibit correlated expression within a microarray dataset. This method comprises the identification of gene sets representing a biological process of interest, determining the subset of genes that exhibit significantly correlated/anti-correlated expression, followed by the quantitative cumulated expression of the set of correlated genes. We applied the pathway index approach by using available gene lists, Pearson correlation, and mean of correlated gene expression values to estimate the pathway status. We then annotated a renal cancer dataset with pathway index scores of 183 canonical pathways, to identify hypoxia deregulation in Renal Cell Carcinoma samples. Using the pathway index approach, we also established a biological correlation between previously identified chromosomal instability signature and proliferation signature in human cancer, and revealed relationships between two distinct prognostic breast cancer signatures, Mammaprint, Oncotype DX, and basal subtypes within a breast tumor dataset. Our analysis indicated that both Mammaprint and Oncotype DX are good prognostic biomarkers and both biomarkers correlated with basal signature score. However, we also demonstrated that both Mammaprint and Oncotype Dx predicted patient outcome in non-basal type cohort respectively, which suggested that in addition to identify basal subtype as bad outcome group, both Mammaprint and Oncotype DX were good prognostic biomarkers in non-basal type cohort. We also applied this approach to analyze genes collinear along chromosome 7 to identify EGFR locus of chromosomal amplification using only transcriptome microarray data. Thus, gene index methodology provides a new tool to annotate microarray data with biological pathway information. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2000.