La méthylation de l'ADN permet, via des remodelages de la chromatine et le recrutement de diverses protéines partenaires, de réguler l'expression des gènes. Des défaillances dans ces mécanismes de régulation peuvent modifier la susceptibilité individuelle face à certaines pathologies, notamment cardiovasculaires. Bien que les différents types cellulaires puissent avoir différents profils de méthylation, l'utilisation de l'ADN provenant de cellules sanguines permet de découvrir de nouveaux mécanismes physiopathologiques. Ce projet de thèse porte sur l'intérêt des analyses d'association méthylome entier comme stratégie alternative aux études d'association génome entier ("GWAS " en anglais) pour identifier de nouveaux déterminants moléculaires de biomarqueurs du risque cardiovasculaire. Pour cela, j'avais à ma disposition deux études épidémiologiques rassemblant 573 sujets pour lesquels les niveaux de méthylation de l'ADN issus du sang périphérique ont été mesurés par une puce à ADN de haute densité couvrant plus de 300 000 sites CpG.Le premier travail que j'ai réalisé a consisté en une étude du méthylome sanguin pour identifier des profils de méthylation associés à l'indice de masse corporelle. Cette étude a permis d'identifier des marques de méthylation de l'ADN au sein du gène HIF3A dont les augmentations sont associées à une augmentation de l'indice de masse corporelle (Lancet, 2014. 383(9933):1990-8). Ces résultats suggèrent en outre que des perturbations de la voie métabolique du gène HIF3A pourraient avoir un rôle important dans la réponse biologique à l'augmentation du poids. Dans un second travail (J Lipid Res, 2014. 55(7):1189-1191), j'ai montré que la variabilité inter individuelle des niveaux de méthylation sanguin du gène CPT1A était associée à la variabilité des taux lipidiques plasmatiques. Ce travail démontre qu'il est possible de détecter à partir d'échantillons sanguins des marques de méthylation de l'ADN qui pourraient être le reflet de mécanismes épigénétiques plus spécifiques de certains types cellulaires ou de certains tissus. Le gène CPT1A est par exemple principalement exprimé dans le foie.Au cours de mon travail de thèse, j'ai également étudié l'influence de la variabilité génétique sur les niveaux de méthylation de l'ADN sanguin (Am J Hum Genet, 2015. 96(4):532-42, Nat Commun, 2015. 6:6326). Cette étude a permis d'identifier près de 3 milles gènes dont les niveaux de méthylation sont associés à la présence de polymorphismes génétiques, localisés soit au sein de ces mêmes gènes (c.-à-d. effet cis) soit à une très grande distance (plus d'une mégabase voire sur un autre chromosome) (c.-à-d. effet trans). Ces résultats ouvrent de nouvelles perspectives pour mieux appréhender la régulation transcriptionnelle de diverses voies métaboliques. / DNA methylation regulates gene expression by chromatin reshaping and the recruitment of various partner proteins. Dysregulation in these regulatory mechanisms can influence the individual susceptibility to some pathologies, including cardiovascular disorders. Although different cell types can have different methylation patterns, the use of DNA from blood cells has recently been proposed as an interesting tool to discover new epigenetic related pathophysiological mechanisms. This PhD project focuses on the interests of the methylome-wide association analyses as an alternative strategy to the fashion genome-wide association studies ("GWAS ") approach to identify new molecular determinants of cardiovascular risk biomarkers. For my project, I had access to two epidemiological studies collecting together 573 subjects in which DNA methylation levels from peripheral blood cells were measured by a high density DNA microarray that covers more than 300 000 CpG sites.The first work I conducted consisted in a study of blood methylome to identify methylation profiles associated with body mass index. This study led to the identification of DNA methylation marks at the HIF3A gene whose increases are associated with an increase in body mass index (Lancet, 2014. 383(9933):1990-8). These results suggest that a disruption of the metabolic pathway HIF3A gene could have an important role in the biological response to the increase of the weight. In a second work (J Lipid Res, 2014. 55(7):1189-1191), I showed that the inter-individual variability in CPT1A methylation levels in blood were associated with variability of plasma lipid levels. This work demonstrates that it is possible to detect DNA methylation marks from blood samples that could reflect epigenetic mechanisms that occur primarily in specific cells or tissues. The CPT1A gene is for example mainly expressed in the liver.During my PhD, I also studied the influence of the genetic variability on the methylation levels from blood DNA (Am J Hum Genet, 2015. 96(4):532-42, Nat Commun, 2015. 6:6326). This work has identified nearly 3000 genes whose methylation levels are associated with the presence of genetic polymorphisms, located either within these same genes (ie cis effect) or at a very large distance (more than one megabase or to another chromosome) (ie trans effect). These results open new perspectives to better understand the transcriptional regulation of various metabolic pathways.

No description available.

京都大学 / マギル大学 / 新制・課程博士 / 博士(ゲノム医学) / 甲第24203号 / 医博JD第1号 / 新制||医||JD1(附属図書館) / 京都大学大学院医学研究科京都大学マギル大学ゲノム医学国際連携専攻 / (主査)教授 稲垣 暢也, 教授 YOUSSEFIAN Shohab, 准教授 Majewski Jacek (マギル大学), 准教授 Gravel Simon (マギル大学), 教授 Gagneur Julien (ミュンヘン工科大学) / 学位規則第4条第1項該当 / Doctor of Philosophy in Human Genetics / Kyoto University / McGill University / DFAM

No description available.

<p>Advancements in human genomic technology have helped to improve our understanding of how genetic variation plays a central role in the mechanism of disease susceptibility. However, the very high dimensional nature of the data generated from large-scale genetic association studies has limited our ability to thoroughly examine genetic interactions. A prioritization scheme – Variance Prioritization (VP) – has been developed to select genetic variants based on differences in the quantitative trait variance between the possible genotypes using Levene’s test (Pare et al., 2010). Genetic variants with Levene’s test p-values lower than a pre-determined level of significance are selected to test for interactions using linear regression models. Under a variety of scenarios, VP has increased power to detect interactions over an exhaustive search as a result of reduced search space. Nevertheless, the use of Levene’s test does not take into account that the variance will either monotonically increase or decrease with the number of minor alleles when interactions are present. To address this issue, I propose a maximum likelihood approach to test for trends in variance between the genotypes, and derive a closed-form representation of the likelihood ratio test (LRT) statistic. Using simulations, I examine the performance of LRT in assessing the inequality of quantitative traits variance stratified by genotypes, and subsequently in identifying potentially interacting genetic variants. LRT is also used in an empirical dataset of 2,161 individuals to prioritize genetic variants for gene-environment interactions. The interaction p-values of the prioritized genetic variants are consistently lower than expected by chance compared to the non-prioritized, suggesting improved statistical power to detect interactions in the set of prioritized genetic variants. This new statistical test is expected to complement the existing VP framework and accelerate the process of genetic interaction discovery in future genome-wide studies and meta-analyses.</p> / Master of Health Sciences (MSc)

Die Methoden der Netzwerktheorie erfreuen sich wachsender Beliebtheit, da sie die Darstellung von komplexen Systemen durch Netzwerke erlauben. Diese werden nur mit einer Menge von Knoten erfasst, die durch Kanten verbunden werden. Derzeit verfügbare Methoden beschränken sich hauptsächlich auf die deskriptive Analyse der Netzwerkstruktur. In der hier vorliegenden Arbeit werden verschiedene Ansätze für die Inferenz über Prozessen in komplexen Netzwerken vorgestellt. Diese Prozesse beeinflussen messbare Größen in Netzwerkknoten und werden durch eine Menge von Zufallszahlen beschrieben. Alle vorgestellten Methoden sind durch praktische Anwendungen motiviert, wie die Übertragung von Lebensmittelinfektionen, die Verbreitung von Zugverspätungen, oder auch die Regulierung von genetischen Effekten. Zunächst wird ein allgemeines dynamisches Metapopulationsmodell für die Verbreitung von Lebensmittelinfektionen vorgestellt, welches die lokalen Infektionsdynamiken mit den netzwerkbasierten Transportwegen von kontaminierten Lebensmitteln zusammenführt. Dieses Modell ermöglicht die effiziente Simulationen verschiedener realistischer Lebensmittelinfektionsepidemien. Zweitens wird ein explorativer Ansatz zur Ursprungsbestimmung von Verbreitungsprozessen entwickelt. Auf Grundlage einer netzwerkbasierten Redefinition der geodätischen Distanz können komplexe Verbreitungsmuster in ein systematisches, kreisrundes Ausbreitungsschema projiziert werden. Dies gilt genau dann, wenn der Ursprungsnetzwerkknoten als Bezugspunkt gewählt wird. Die Methode wird erfolgreich auf den EHEC/HUS Epidemie 2011 in Deutschland angewandt. Die Ergebnisse legen nahe, dass die Methode die aufwändigen Standarduntersuchungen bei Lebensmittelinfektionsepidemien sinnvoll ergänzen kann. Zudem kann dieser explorative Ansatz zur Identifikation von Ursprungsverspätungen in Transportnetzwerken angewandt werden. Die Ergebnisse von umfangreichen Simulationsstudien mit verschiedenstensten Übertragungsmechanismen lassen auf eine allgemeine Anwendbarkeit des Ansatzes bei der Ursprungsbestimmung von Verbreitungsprozessen in vielfältigen Bereichen hoffen. Schließlich wird gezeigt, dass kernelbasierte Methoden eine Alternative für die statistische Analyse von Prozessen in Netzwerken darstellen können. Es wurde ein netzwerkbasierter Kern für den logistischen Kernel Machine Test entwickelt, welcher die nahtlose Integration von biologischem Wissen in die Analyse von Daten aus genomweiten Assoziationsstudien erlaubt. Die Methode wird erfolgreich bei der Analyse genetischer Ursachen für rheumatische Arthritis und Lungenkrebs getestet. Zusammenfassend machen die Ergebnisse der vorgestellten Methoden deutlich, dass die Netzwerk-theoretische Analyse von Verbreitungsprozessen einen wesentlichen Beitrag zur Beantwortung verschiedenster Fragestellungen in unterschiedlichen Anwendungen liefern kann.

A doença de Chagas é causada pelo protozoário Trypanosoma cruzi e atinge cerca de 12 milhões de pessoas no continente americano, a forma clássica de transmissão ocorre por intermédio do inseto vetor da subfamília Triatominae, popularmente chamado de barbeiro.Em um trabalho anterior, foi realizada uma análise de segregação complexa que indicou a presença de um gene principal, com um componente multifatorial influenciando a predisposição à infecção por Trypanosoma cruzi. A população é composta por 4697 indivíduos pertencentes a 886 famílias vindas do Nordeste do Brasil e tiveram os dados e amostras de sangue e saliva coletados entre 1969 e 1970.No presente estudo foi utilizada uma amostra de 69 indivíduos, sendo 18 positivos para a infecção por Trypanosoma cruzi e 51 negativos, distribuídos em 14 famílias. Os indivíduos tiveram seu DNA extraído e genotipado utilizando microarranjos de DNA de 260 K SNPs (GeneChip Mapping Affymetrix). Testes de associação mostraram significância entre a infecção por T. cruzi e o SNP rs17469997 do cromossomo 10, com P=0,015 após a correção de Bonferroni. Para validar estes inéditos resultados, análises de ligação multi-ponto foi feita com o programa GeneHunter (KRUGLYAK et al., 1996) e ligação dois-pontos com o programa SuperLink (FISHELSON e GEIGER, 2002), mas ambas não apresentaram resultados significativos, devido ao pequeno número de famílias informativas. / Chagas disease is caused by the protozoan Trypanosssoma cruzi and is usually transmitted by Triatominae bugs and affects about 12 million people in the American continent. In a previous study, segregation analysis showed evidence of a major gene with a small multifactorial component influencing the predisposition to the Trypanosoma cruzi infection in a population composed by 4697 individuals of 886 families from Northeastern Brazil in 1969-1970 at São Paulo, Brazil. In the present work, 69 individuals (18 positives to T. cruzi infection and 51 negative) belonging to 14 families were selected. They had the DNA extracted and genotyped using 250K SNPs DNA microarrays (GeneChip Mapping Affymetrix). 18 SNPs showed evidence of association between infection to T. cruzi with P<10-5, although after Bonferroni\'s correction only the SNP rs17469997 (minor allele frequency = 0.1667, adjusted-Bonferroni P = 0.015) on chromosome 10 was significant. The other 17 SNPs that showed association with T. cruzi infection with P<10-5 can still be informative in linkage analyses. On an effort to validate these findings, a multi point linkage analyses was performed with GeneHunter (KRUGLYAK et al., 1996) program and a two point linkage analyses were performed with SuperLink (FISHELSON e GEIGER, 2002) program, both analyses showed no significant results.

[Truncated abstract] This thesis investigates novel methods for the genetic association analysis of haplotype data in samples of unrelated individuals, and applies these methods to the analysis of coronary heart disease and related phenotypes. Determining the inheritance pattern of genetic variants in studies of unrelated individuals can be problematic because family members of the studied individuals are often not available. For the analysis of individual genetic loci, no problem arises because the unit of interest is the observed genotype. When the unit of interest is the linear combination of alleles along one chromosome, inherited together in a haplotype, it is not always possible to determine with certainty the inheritance pattern, and therefore statistical methods to infer these patterns must be adopted. Due to genotypic heterozygosity, mutliple possible haplotype configurations can often resolve an individual's genotype measures at multiple loci. When haplotypes are not known, but are inferred statistically, an element of uncertainty is thus inherent which, if not dealt with appropriately, can result in unreliable estimates of effect sizes in an association setting. The core aim of the research described in this thesis was to develop and implement a general method for haplotype-based association analysis using multiple imputation to appropriately deal with uncertainty haplotype assignment. Regression-based approaches to association analysis provide flexible methods to investigate the influence of a covariate on a response variable, adjusting for the effects of other variables including interaction terms. ... These methods are then applied to models accommodating binary, quantitative, longitudinal and survival data. The performance of the multiple imputation method implemented was assessed using simulated data under a range of haplotypic effect sizes and genetic inheritance patterns. The multiple imputation approach performed better, on average, than ignoring haplotypic uncertainty, and provided estimates that in most cases were similar to those observed when haplotypes were known. The haplotype association methods developed in this thesis were used to investigate the genetic epidemiology of cardiovascular disease, utilising data for the cholesteryl ester transfer protein gene (CETP), the hepatic lipase (LIPC) gene and the 15- lipoxygenase (ALOX15) gene on a total of 6,487 individuals from three Western Australian studies. Results of these analyses suggested single nucleotide polymorphisms (SNPs) and haplotypes in the CETP gene were associated with increased plasma high-density lipoprotein cholesterol (HDL-C). SNPs in the LIPC gene were also associated with increased HDL-C and haplotypes in the ALOX15 gene were associated with risk of carotid plaque among individuals with premature CHD. The research presented in this thesis is both novel and important as it provides methods for the analysis of haplotypic associations with a range of response types, while incorporating information about haplotype uncertainty inherent in populationbased studies. These methods are shown to perform well for a range of simulated and real data situations, and have been written into a statistical analysis package that has been freely released to the research community.

No description available.

[Truncated abstract] Background: Genetic epidemiology draws on the mechanisms of heredity and the reproductive characteristics of populations to formulate methods to investigate the role of genetic factors and their interaction with the environment in disease aetiology. Asthma and atopy are complex genetic disorders and are among the most common diseases to affect the developed world. Twin studies provide an elegant means of disentangling genetic and environmental contributions to the aetiology of conditions that have a significant impact on the health of the general population in ways that cannot be achieved by any other study design, by comparing disease frequency in monozygotic (MZ) or identical twins, who share 100% of their genes with that in dizygotic (DZ) or non-identical twins who share, on average, 50% of their genes. Twin-family studies allow the complete partitioning of phenotypic variation into components representing additive genetic, dominance, shared environment and non-shared environment. ... For twin family data, the best fitting model was the one which included additive genetic effects and either genetic dominance or shared sibling environment, and that shared family environment was not important. With respect to asthma in WA twin families, there are no reasons to conclude that the EEA is not valid. Conclusions: The WA Twin Register is the first population-based register of childhood multiples to be established in Australia, and the WATCH study is one of only a few population-based twin-family studies in the world. Families who participated in the WATCH study were no different from non-participants with respect to social class and there was no difference in the prevalence of DDA in WATCH study twins and either their singleton siblings or the general population of WA children. Results from the GEE models replicate those found in numerous studies from many different countries. The BUGS models developed have been shown to produce consistent results with both simulated and real data sets and offer alternative methods of analyzing twin and twin-family data. By including an extra term in the partitioning of the variance to account for the environment effect of being a MZ twin, a numerical value is calculated for the difference in MZ and DZ correlation with respect to the phenotype examined, which allows the validity of the EEA to be directly assessed.