An integrated experimental and computational approach to understanding complex disease: Differential gene expression in type 2 diabetes mellitus.

In this work, I focus on the development and testing of tools to explore differential gene expression as a molecular mechanism underlying the predisposition to complex disease. By definition, complex diseases (e.g. hypertension, diabetes, asthma) demonstrate both environmental and genetic factors in disease susceptibility, implicating genes that show both environmental response and genetic variability in predisposing disease. Differential gene expression is one important cellular response that reflects both of these factors in complex disease. Given their combined genetic and environmental influences, complex diseases are studied using a variety of analytical approaches. Consequently, the integration of data from diverse sources is a significant and important challenge in modern biomedical science. I demonstrate a statistical approach to integrating differential gene expression and genetic linkage data from multiple sources to reveal candidate genes, metabolic pathways, and DNA regulatory elements related to one complex disease. Using Type 2 Diabetes Mellitus (T2DM) as a sample complex disease, I show that using combined criteria for differential gene expression and genetic linkage yields a better selection of candidate genes than using either criterion alone. Based on this integrated approach, I identify four candidate T2DM predisposing genes, two of which are novel. I next show that differential expression among groups of genes working in metabolic pathways provides insight into the roles of these metabolic pathways in predisposing disease. I identify retinol metabolism as a candidate metabolic pathway in T2DM predisposition. Finally, genetic variation in transcription factor binding motifs may lead to variation in gene expression and, for disease related genes, variable predisposition to disease. Using a simple cluster of co-expressed T2DM related genes, I show that these transcription factor binding motifs can be identified by evolutionary and functional conservation of regulatory DNA sequences in the promoters of genes that are differentially expressed in response to disease-related environmental influences. The analytical strategies and techniques developed in this work are effective when applied to T2DM. In addition, they depend on characteristics common to all complex diseases, suggesting that they could be applied to finding candidate genes, metabolic pathways, and DNA regulatory elements implicated in other complex diseases.PhDBiological SciencesBiostatisticsGeneticsUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/124505/2/3150040.pd