Until date traditional genetic methods for disease gene identification have been successful largely for single gene diseases. However, the elucidation of genetic factors underlying diseases with complex etiology, where multiple loci affect susceptibility and/or penetrance, and the environment may play a role also, has been less successful. These diseases, including atherosclerosis, diabetes, and obesity, are the largest sources of 'natural' morbidity in Western society. To this end, we have developed several systems biology based methods to elucidate the complex genetics underlying these diseases. These methods often make use of the analysis of a segregating mouse population. In such a population one can gather clinical trait data, genetic marker data, and gene expression data and track the co-segregation of these three variables throughout the population under study. By building on this basic premise, we have developed methods which allow us to propose candidate genes for clinical traits, model the transcriptional networks associated with these traits, and propose genes acting as key drivers in these complex networks. We have used these novel methods to produce several candidate genes for complex traits such as obesity and metabolic syndrome. We have shown that polymorphic alleles within genes often affect their expression, and also can affect the expression of other genes. We have explored the potential for these genes to be candidate genes underlying clinical trait quantitative trait loci. We have also constructed a liver gene co-expression network and found gene modules which are highly related to obesity. We then used a novel strategy which combines the use of genetics and network properties to model gene/trait relationships. In summary, we have contributed methodology involving novel systems level analyses which allow the scientific community to better understand the complex genetics underlying obesity and obesity-related traits.
展开▼
