Typescript (photocopy) / 157 leaves : / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Thesis (M.D.)--Dept. of Medicine, University of Adelaide, 1983

Thesis (M.D.) - Dept. of Medicine, University of Adelaide, 1983. / Typescript (photocopy).

Genome-wide association (GWA) studies have been successfully applied in detection of susceptibility loci for complex diseases, but most of the identified variants have a large to moderate effect, and explain only a limited proportion of the heritability of the diseases. It is believed that the majority of the latent risk alleles have very small risk effects that are difficult to be identified and GWA study may have inadequate power in dealing with those small effect variants. Researchers will often collect other phenotypic information in addition to disease status to maximize the output from the study. Some of the phenotypes can be on the pathway to the disease, i.e., intermediate phenotype. Statistical methods based on both the disease status and intermediate phenotype should be more powerful than a case-control study as it incorporates more information. Meta-analysis has been used in genetic association analysis for many years to combine information from multiple populations, but never been used in a single population GWA study. In this study, simulations were conducted and the results show that when an intermediate phenotype is available, the meta-analysis incorporating the disease status and intermediate phenotype information from a single population has more power than a case-control study only in GWA study of complex diseases, especially for identification of those loci that have a very small effect. And compared with Fisher's method, the modified inverse variance weighted meta-analysis method is more robust as it is more powerful and has a lower type I error rate at the same time, which provides a potent approach in detecting the susceptibility loci associated with complex diseases, especially for those latent loci whose effect are very small. In the meta-analysis of lung cancer with smoking data, the results replicate the signal in \emph{CHRNA3} and \emph{CHRNA5} genes on chromosome 15q25. Some new signals in \emph{CYP2F1} on chromosome 19, \emph{SUMF1} on chromosome 3, and \emph{ARHGAP10} on chromosome 4 are also detected. And the \emph{CYP2F1} gene, close to the already known cigarette-induced lung cancer gene \emph{CYP2A6}, is highly possible another cytochrome P450 (CYP) gene that is related to the smoking-involved lung cancer. The meta-analysis of rheumatoid arthritis with anti-cyclic citrullinated peptide (anti-CCP) data identified new signals on 9q24 and 16q12. There are evidences these two regions are involved in other autoimmune diseases and different autoimmune/inflammatory diseases may share same genetic susceptibility loci. Both the theoretical and empirical studies show that the modified variance weighted meta-analysis method is a robust method and is a potent approach in detecting the susceptibility loci associated with complex diseases when an intermediate phenotype is available.

Many human traits and diseases have a polygenic architecture, where phenotype is partially determined by variation in many genes. These complex traits or diseases can be highly heritable and genome-wide association studies (GWAS) have been relatively successful in the identification of associated variants. However, these variants typically do not account for most of the heritability and thus, the genetic architecture remains uncertain.

国立情報学研究所で電子化したコンテンツを使用している。

Thesis (Ph. D.)--University of Adelaide, 1986. / Includes bibliographical references (leaves 119-136).

Cette thèse a pour objet l'étude du lien génotype-transcriptome et de son influence sur le développement des maladies multifactorielles. Les apports de ce travail sont à la fois méthodologiques et appliqués. Nous étudions d'abord le lien génotype-transcriptome en établissant la liste des eQTL (expression Quantitative Trait Loci) dans le monocyte et nous évaluons l'apport de l'observation des eQTL pour l'interprétation des analyses d'association génome entier (GWAS). Nous proposons ensuite une méthode pour l'identification de variants génétiques affectant des modules de gènesco-régulés que nous appliquons à l'étude des données d'expression de monocytes issus d'une large étude populationnelle (GHS). Nous mettons ainsi en évidence plusieurs loci affectant l'expression de modules de gènes co-régulés, dont plusieurs sont impliqués dans la prédisposition au diabète de type I. Nous montrons également que le processus d'isolation des cellules monocytaires peut engendrer des biais liés à la contamination par des types cellulaires non désirés et nous proposons une approche pour contrôler ce type de biais dans les analyses. / This thesis deals with the study of the relation between genotype and expression and its influence on the development of complex diseases. This work brings both methodological and applied results.First, we study the relation between genotype and transcriptome by establishing a database of eQTL (expression quantitative Trait Loci) in monocytes and we evaluate the contribution of eQTL for the interpretation of results from Genome Wide Association Studies (GWAS).We then provide a methodology for the identi_cation of genetic polymorphisms regulating modules of co-expressed genes that we apply to a large scale populationnal study of the monocyte transcriptome.We thus identify several loci associated with modules of co-regulated genes, several of which are involved in the susceptibility to type I diabetes. We also show that the isolation of monocytes can induce complex bias through contamination from unwanted cell types and we provide a method to control for such bias in the analysis.

Bibliography: leaves 119-136.

<p>Systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) are complex autoimmune diseases affecting 0.05-2% of the population worldwide. </p><p>Genetic studies detected linkage with SLE in the 2q37 region, and intensive family-based and case-control association studies in several populations identified that allele A of the SNP PD-1.3 located in the immunoreceptor PDCD1 (PD-1) gene, increases risk of the disease by 2.6-fold in Caucasians (p<0.00001) and by 3.5-fold in Mexicans (p=0.0009). </p><p>The same allele was found to be a risk factor for lupus nephritis, a severe clinical manifestation of SLE. In Swedish and European-American females with SLE, patients with the allele A had nephritis 1.8 times (p=0.01) more often than patients with allele G .</p><p>Moreover, the allele A was also found 1.8 times (p=0.005) more often in RA patients, negative for the known risk-factors, rheumatoid factor and the shared epitope, than in other groups of patients and controls. </p><p>Functional studies demonstrated that the mechanism behind the SNP PD-1.3 is related to the disruption of the binding site for RUNX transcription factors in the regulatory region. Expression of the PD-1 and RUNX genes was altered in the activated T cells of SLE patients compared to controls.</p><p>The Tumor Necrosis Factor Receptor 2 (TNFR 2) gene was studied as a second candidate gene for both SLE and RA. The results of our studies in SLE and RA patients and controls from Sweden and Mexico do not support the association of the polymorphism TNFR 2 M196R with these diseases. Other polymorphisms in this gene and other genes in this region should therefore be studied.</p>

Systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) are complex autoimmune diseases affecting 0.05-2% of the population worldwide. Genetic studies detected linkage with SLE in the 2q37 region, and intensive family-based and case-control association studies in several populations identified that allele A of the SNP PD-1.3 located in the immunoreceptor PDCD1 (PD-1) gene, increases risk of the disease by 2.6-fold in Caucasians (p&lt;0.00001) and by 3.5-fold in Mexicans (p=0.0009). The same allele was found to be a risk factor for lupus nephritis, a severe clinical manifestation of SLE. In Swedish and European-American females with SLE, patients with the allele A had nephritis 1.8 times (p=0.01) more often than patients with allele G . Moreover, the allele A was also found 1.8 times (p=0.005) more often in RA patients, negative for the known risk-factors, rheumatoid factor and the shared epitope, than in other groups of patients and controls. Functional studies demonstrated that the mechanism behind the SNP PD-1.3 is related to the disruption of the binding site for RUNX transcription factors in the regulatory region. Expression of the PD-1 and RUNX genes was altered in the activated T cells of SLE patients compared to controls. The Tumor Necrosis Factor Receptor 2 (TNFR 2) gene was studied as a second candidate gene for both SLE and RA. The results of our studies in SLE and RA patients and controls from Sweden and Mexico do not support the association of the polymorphism TNFR 2 M196R with these diseases. Other polymorphisms in this gene and other genes in this region should therefore be studied.