Doctor of Philosophy, Case Western Reserve University, 2010, Genetics

Genetics has been invaluable in advancing the understanding of complex trait and disease inheritance, providing insight to the genes and pathways underlying heritable phenotypes. However, the variants contributing to many highly heritable human diseases have remained largely undetectable or ‘missing' in biomedical research. Many possibilities exist for why the causative variants may remain undetected including the additive effects of many genes with small effects, overestimates of heritability, epistatic interactions between variants and epigenetic inheritance. The goal of my project was to characterize transgenerational genetic effects, resulting from genetic variants in previous generations that are not inherited in the at-risk individuals, and their role in the inheritance of complex phenotypes. This project involved two components. First, I examined the frequency and magnitude of transgenerational effects using Y-chromosome substitution strains (CSSs) to generate genetically identical females with genetically distinct parents and found that transgenerational genetic effects occur at similar rates and with similarly large effects compared to conventional risk loci in CSSs. I concluded that transgenerational effects contribute to phenotypic inheritance at an appreciable frequency and magnitude. Second, I identified Apobec1 as a novel genetic modifier of testicular cancer risk and characterized the lineage-specific and transgenerational effects of Apobec1, an RNA-editing enzyme, on the prevalence of testicular germ cell tumors (TGCTs) and efficiency of stem cell derivation. Finally, I identified transgenerational genetic epistasis between Apobec1 deficiency and Dnd1Ter, an established TGCT risk enhancer, affecting TGCT prevalence and allele segregation in offspring. I concluded that RNA biology, and possibly RNA-editing, play a role in TGCT pathogenesis and possibly in mediating transgenerational inheritance.

Committee: Joseph H Nadeau PhD (Advisor); Peter Scacheri PhD (Committee Chair); Georgia Wiesner MD (Committee Member); Angela Ting PhD (Committee Member); Ruth Keri PhD (Committee Member) Subjects: Genetics

Doctor of Philosophy, Case Western Reserve University, 2010, Epidemiology and Biostatistics

Complex disease sometimes has system-wide level impact by affecting a constellation of physiologically interrelated phenotypes rather than a single phenotype, resulting in a set of co-morbidities. The physiological connection between the phenotypes, and the consequent co-occurrence of morbidities, varies from individual to individual in a way that can affect diseases prognosis. As an example, when obesity and its associated morbidities, dyslipidemia, hypertension and insulin resistance, do co-occur, this co-occurrence increases risk of diabetes and coronary heart disease in a way not explained by the presence of each individual morbidity alone. In this work, the physiological connections for each individual are characterized by the correlation values that the phenotypes present in their repeated measurements throughout the individual's life. The variation in the within-individual phenotypic relationships, from individual to individual, can then be studied as a new quantitative trait. First, this study shows that traditional genetic approaches which target variation in the phenotypic values do not capture the variation in within individual phenotypic correlations. Secondly, two approaches designed to specifically model the new quantitative trait are statistically compared. Finally, the biological relevance of the phenotypic correlations underlying obesity and its associated morbidities is investigated using the Framingham heart study data (human data) and the C57BL/6J and A/J chromosome substitution strain panel (mouse data). It is found that these phenotypic correlations are associated to diabetes and cardiovascular disease in a way not explained by the phenotypic values alone. It is also shown that there is genetic variation underlying these phenotypic correlations and that it is distinct and independent from that underlying the phenotypic values. This work concludes that approaches that exclusively model phenotypic values when studying the genetics of co-morbidi (open full item for complete abstract)

Committee: J. Sunil Rao PhD (Committee Chair); Joseph Nadeau PhD (Advisor); Xiaofeng Zhu PhD (Committee Member); Catherine Stein PhD (Committee Member) Subjects: Biomedical Research; Biostatistics; Genetics