Integrating genes, function and phenotype to dissect the genetic architecture of complex diseases

Dissecting the genetic basis of complex diseases presents a significant challenge, given that they are the result of a complex interplay between multiple genetic and environmental factors. Genome-wide association studies (GWAS) have been the predominant approach used in the search for genetic determinants of susceptibility to complex diseases. Although GWAS have identified many variants associated with numerous diseases, they explain only a fraction of the heritability and little is known about their functional consequences. A major challenge to GWAS in identifying genetic risk factors stems from the fact that genetic susceptibility to complex diseases is dependent on context, such as environment, lifestyle and host factors. Thus, we hypothesize that incorporation of biologically relevant context into GWAS can help elucidate the genetic basis and biology of complex diseases. Here we test this hypothesis using two separate approaches. First, we investigate the genetic susceptibility to melanoma, in the context of sunlight exposure, the major causal factor for melanoma. Second, we investigate the genetic susceptibility to inflammatory and malignant colonic diseases in the context of tissue-specific function and ancestry, by studying gene expression variation in the African American colon.;Melanoma is the fastest growing cancer in the United States, and presents a significant health care burden. The majority of melanoma cases are attributable to ultraviolet radiation (UVR) exposure. Lighter skin color is a risk factor for melanoma; this is due to the protective role of melanin, which absorbs UVR and prevents it from reaching deeper layers of the skin to cause DNA damage. These observations suggest that genetic architecture of melanoma may be dependent on the context of skin color, which is a modifier of realized doses of UVR in the skin. To test this hypothesis, we estimated the heritability of melanoma, stratifying on skin color in a cohort of 1,888 cases and 990 controls of European ancestry. We found that the heritability of melanoma is significantly higher in dark-skinned whites compared to light-skinned whites. We further investigated the genetic architecture of melanoma, through the prism of heritability, and showed that risk loci are concentrated in a handful of genomic regions with relatively large effects. Taken together, these results suggest that the genetic contribution to melanoma risk is significantly greater in dark-skinned whites than in light-skinned whites, and that GWAS of melanoma can gain significantly in power, by incorporating skin color in the analysis. More broadly, our results suggest that heritability can be used as a useful tool to guide genetic association study design by identifying patient subsets, who may share a similar genetic etiology.;In the second part of this thesis, we investigated genetic susceptibility to colonic diseases in the context of tissue-specific function and ancestry. GWAS have identified numerous risk variants for inflammatory bowel disease as well as colorectal cancer. However, the genes and mechanisms underlying susceptibility have not yet been elucidated and resources specific to the colon are lacking. Furthermore, there are inter-ethnic differences in colonic disease susceptibility between European and African Americans, but their genetic basis is unknown. Therefore, we performed comprehensive eQTL mapping in the African American colon and validated its utility as a resource for genetic studies of colonic disease. Our study provides a tissue-specific tool to improve understanding of the biology underlying genetic susceptibility to colonic diseases, as well as inter-ethnic differences in colonic disease risk.;Our study provides unique insights into the genetic susceptibility to melanoma and colonic diseases and demonstrates the successful incorporation of context in GWAS to elucidate complex disease genetics and biology. These findings will guide future studies aimed at identifying susceptibility genes for complex diseases.