No description available.

Salt stress is a global problem that limits agricultural production. The early responses to salinity, independent of toxic shoot-ion accumulation, are still largely unknown. Here, optimised salt treatment and high-throughput phenotyping protocols were developed and used to examine the natural variation in the early responses to salt stress of 191 Arabidopsis thaliana accessions. Common and novel traits of plants grown under salt treatment were captured through time using RGB and chlorophyll fluorescence imaging. Phenotypic data was combined with the Arabidopsis 10M SNP markers for genome-wide association studies to identify genetic components underlying the early responses to salt stress. The most promising candidate loci were selected based on association strength, allele frequency and number of traits associating to the same locus. In silico analysis highlighted interesting allelic variations across the identified loci, and by phenotypically characterising the candidate gene mutants under salt stress, the associations were experimentally validated. This work comprises a detailed study of the natural variation in the early responses to salt stress, which can give insights into the mechanisms contributing to salinity tolerance and provide the fundaments for crop improvements under saline conditions across the globe.

Poliomyelitis is a foremost cause of paralysis among preventable diseases among children and adolescents globally. It is caused by persistent infection with poliovirus (PV). The PV infection does not always cause paralysis. A lack of immunization always increases the risk of paralytic polio. Genetic factors also been shown to affect the risk of developing the disease. The aim of this thesis is to investigate whether there are any genetic associations to paralytic poliomyelitis. This is based on a model for understanding its nature as a complex disease, where many genes are involved in contributing to the disease state. This is a population-based case-control study to identify genetic loci that influence disease risk. The study examined the association of genetic variation in single nucleotide polymorphisms (SNPs) across the genome with paralytic poliomyelitis susceptibility in the United States and Canadian survivors of poliomyelitis population, using a genome-wide association study (GWAS) approach. No association was observed. Loci that have been previously implicated were not found to affect the susceptibility to poliomyelitis in this study. The thesis consists of four chapters. Chapter 1 describes the epidemiology, pathogenesis and management of poliomyelitis. Chapter 2 gives an overview of the genomics of infectious diseases in general. Chapter 3 introduces the study population and presents the genome-wide analysis and associations with logistic regression to identify loci explore genes that might be associated with paralytic poliomyelitis and presents results. Chapter 4 discusses the implications of the results and explains future directions. / Thesis / Master of Science (MSc)

Genome-wide association studies (GWAS) seek to identify single nucleotide polymorphisms (SNP) causing phenotypic responses in individuals. Commonly, GWAS analyses are done by using single marker association testing (SMA) which investigates the effect of a single SNP at a time and selects a candidate set of SNPs using a strict multiple correction penalty. As SNPs are not independent but instead strongly correlated, SMA methods lead to such high false discovery rates (FDR) that the results are difficult to use by wet lab scientists. To address this, this dissertation proposes three different novel Bayesian methods: BICOSS, BGWAS, and IEB. From a Bayesian modeling point of view, SNP search can be seen as a variable selection problem in linear mixed models (LMMs) where $p$ is much larger than $n$. To deal with the $p>>n$ issue, our three proposed methods use novel Bayesian approaches based on two steps: a screening step and a model selection step. To control false discoveries, we link the screening and model selection steps through a common probability of a null SNP. To deal with model selection, we propose novel priors that are extensions for LMMs of nonlocal priors, Zellner-g prior, unit Information prior, and Zellner-Siow prior. For each method, extensive simulation studies and case studies show that these methods improve the recall of true causal SNPs and, more importantly, drastically decrease FDR. Because our Bayesian methods provide more focused and precise results, they may speed up discovery of important SNPs and significantly contribute to scientific progress in the areas of biology, agricultural productivity, and human health. / Doctor of Philosophy / Genome-wide association studies (GWAS) seek to identify locations in DNA known as single nucleotide polymorphisms (SNPs) that are the underlying cause of observable traits such as height or breast cancer. Commonly, GWAS analyses are performed by investigating each SNP individually and seeing which SNPs are highly correlated with the response. However, as the SNPs themselves are highly correlated, investigating each one individually leads to a high number of false positives. To address this, this dissertation proposes three different advanced statistical methods: BICOSS, BGWAS, and IEB. Through extensive simulations, our methods are shown to not only drastically reduce the number of falsely detected SNPs but also increase the detection rate of true causal SNPs. Because our novel methods provide more focused and precise results, they may speed up discovery of important SNPs and significantly contribute to scientific progress in the areas of biology, agricultural productivity, and human health.

Forest tree breeding methods rank among the most efficient ways to increase productivity and quality of forests. With the advent of high-throughput genotyping technology, genome-enabled breeding has started to gain importance and may overcome some weaknesses of traditional tree breeding. Genomic Selection (GS), which involves using genome-wide markers to predict breeding values of individuals in a population, has been proposed for animal and plant breeding programs. GS enables very accurate selection decisions through estimation of genomic estimated breeding values (GEBVs). While the goal of GS is to predict phenotype from genotype, it does not identify the underlying genes that have important roles in a trait. Genome-Wide Association Studies (GWAS) approaches are therefore complementary to GS, enabling identification of these genes, which may be useful for marker-assisted selection in some traits. In this study, we first estimated heritability for several adaptive traits (cold hardiness, dbh, bud flush, height, and bud set) in a population of Populus trichocarpa and for height, diameter, and stem straightness in Pinus taeda. GEBVs accuracies were estimated using a ridge regression–best linear unbiased prediction (rrBLUP) model, and these accuracies were compared with estimated heritabilities. GWAS was also performed for the both imputed and non–imputed data of P. taeda population using TASSEL (Trait Analysis by aSSociation Evolution and Linkage) software, as well as rrBLUP and FFBSKAT (Fast Family-Based Sequence Kernel Association Test) packages in R. Heritabilities ranged from 0.34 to 0.56 for P. trichocarpa and 0.14 to 0.37 for P.taeda. GWAS identified 3244 associations for dbh, 4077 associations for stem straightness, and 5280 SNPs for height (p≤0.05) in TASSEL using the reduced model (marker data only), whereas 2729, 3272 and 3531 associations were found with the full model where we also included population structure as a covariate. FFBSKAT showed a similar number of SNP associations (2989, 3046 and 3058). There was an inflation of SNP associations (~20k) found in rrBLUP, which suggests population structure was not effectively controlled. The GEBVs accuracies ranged from 0.09 and 0.22 for P.trichocarpa and 0.09 to 0.23 for P.taeda using rrBLUP method. Testing the effect of repetation on the accuracy of GEBV for poplar showed that there was no significant difference between the number of cycles. Also, there was no significant difference the accuracy of GEBVs in pine between two different imputation methods, the marker mean value and Beagle software. / Master of Science

Chickens are ideal model organism to improve understanding of several research areas as phylogenetic, embryology, biomedicine, livestock, and have recently been suggested as a promising model for epigenetic studies. In the livestock area, chickens are source of protein to humans and had been selected to achieve a high production standards based on genetic breeding by the traditional selection. We are now in the genomics and epigenomics era and it is time be concern about the use of new tools to improve selection not only thinking about production, but also in the health and welfare of animals. The use of molecular approaches, have been a fundamental tool to understand biological models and improve selection strategies based on genomic information in breeding programs. Molecular approaches have also contributed to understanding of the evolutionary history of these models and the genetics and epigenetics mechanisms involved in evolution process and genetic diversification of chickens. In this context, many technologies have emerged to produce high-throughput data using Next-generation sequencing (NGS) approaches. NGS provided a large amount of information for diverse purposes such as to detect single nucleotide polymorphisms (SNPs), and methylated DNA profiles in chickens. In addition, NGS has allowed the development of pre-designed SNP arrays for genome-wide association studies (GWAS) with specific phenotypes of interest. Moreover, although NGS has enough power to detect informative polymorphisms, its high cost makes it impractical to be used in GWAS and Methylated DNA immunoprecipitation sequencing (MeDIPseq) studies. The demand for an economical, efficient, simple-step and reliable genome-wide method of SNPs discovery, validation and characterization, was the reason for the development of this study. We applied reduced representation sequence by restriction enzyme (RE) cleavage of target chicken genome to be applied in GWAS. Thereafter, to combine the reduced representation of the genome with MeDIPseq method, we developed a novel approach to perform differential methylation studies using reduced libraries. These works allowed us to identify SNPs associated with performance traits and differential methylation windows related to different stress conditions in chickens. / A galinha é um organismo modelo ideal para melhorar o entendimento de diversas áreas da pesquisa como: filogenética, embriologia, biomedicina, pecuária, e tem sido recentemente sugerida como um modelo promissor para estudos em epigenética. Na pecuária, as galinhas são fonte de proteína para os seres humanos e tem sido alvo de seleção para alcançar um alto padrão de produção com base no melhoramento genético tradicional. Mas agora, estamos na era genômica e epigenômica e as atenções devem ser voltadas para o uso de novas ferramentas para melhorar a seleção não só pensando em produção, mas também na saúde e bem-estar dos animais. O uso de abordagens moleculares, tem sido uma ferramenta fundamental para compreender modelos biológicos e melhorar as estratégias de seleção baseadas na informação genômica em programas de melhoramento. Abordagens moleculares, também tem contribuído para a compreensão da história evolutiva desses modelos e os mecanismos genéticos e epigenéticos envolvidos no processo de evolução e diversificação genética das galinhas. Neste contexto, tecnologias evoluíram para produção de dados de sequenciamento de alto rendimento por sequenciamento de próxima geração (NGS). NGS forneceu uma grande quantidade de informação a ser utilizado para diversos fins, como para detectar polimorfismos de nucleotídeo único (SNPs) e perfis de metilação diferencial do DNA em galinhas. NGS tem permitido também o desenvolvimento de painéis de SNP para testes de associações genômica ampla (GWAS) com fenótipos específicos de interesse. Embora NGS tem poder suficiente para detectar polimorfismos informativos, o seu elevado custo o torna impraticável para ser utilizado em GWAS ou estudos de metilação diferencial por sequenciamento de DNA metilado por imunoprecipitação (MeDIPseq). A procura de um método de genotipagem eficiente, simples, econômico e confiável para descoberta, caracterização e validação de SNPs, foi a razão para o desenvolvimento deste estudo. Utilizamos sequenciamento do genoma reduzido por enzima de restrição (RE) que cliva o genoma alvo para identificação de SNPs nestas bibliotecas reduzidas e aplicação deste método em GWAS. Em seguida, para combinar a representação reduzida do genoma com o método MeDIPS, desenvolvemos uma nova abordagem para a realização de estudos de metilação diferencial utilizando as bibliotecas reduzidas. Estes trabalhos permitiram a identificação de SNPs associados com características de desempenho e janelas de metilação diferencial relacionados a diferentes condições de manejo em galinhas.

O leite bovino é um alimento essencial na nutrição humana, principalmente para os mais jovens por ser uma fonte importante de proteínas, minerais e vitaminas. A crescente demanda por quantidade e qualidade de leite nos últimos anos tem impulsionado inúmeras pesquisas, principalmente em relação ao perfil de ácidos graxos por diferir substancialmente das exigências humanas. Nutrição e melhoramento genético são os principais fatores capazes de promover a alteração da qualidade nutricional do leite. Por muito tempo as mudanças alcançadas via melhoramento genético foram obtidas exclusivamente com base na genética quantitativa, mas a partir dos anos 90 o interesse nesta área passou a ser divido com a genética molecular. Os avanços alcançados na biologia molecular obtidos por meio do mapeamento e sequenciamento do genoma das diversas espécies e de estudos de associação fenótipo-genótipo vêm auxiliando a explicar o fundo genético das características quantitativas. Atualmente, a maioria dos estudos de associação fenótipo-genótipo foram delineados para estimar apenas efeitos genéticos aditivos em um único lócus. No entanto, parte da variação observada nestes fenótipos resultam da interação entre loci ou genes, tornando estes estudos fundamentais para conhecer a origem da variação biológica destas características. Assim, os objetivos deste estudo foram: I) Associar polimorfismos de nucleotídeo único (SNPs) com características produtivas e o perfil de ácidos graxos no leite; e II) Identificar interações entre os SNPs associados a estas características no leite de bovinos da raça Holandesa. Fenótipo ajustado de 760 vacas para o efeito fixo de grupo de contemporâneo e covariáveis dias em lactação, idade à mensuração e produção de leite, e 6.553 SNPs foram considerados para realizar a associação por meio de regressão SNP a SNP. No total nove SNPs foram associados a uma ou mais características relacionadas ao teor de gordura e perfil de ácidos graxos. Algumas destas associações evidenciam a presença de epistasia e/ou pleiotropia entre estas regiões. Para a identificação de interação entre os SNPs, foram usados modelos que consideraram os efeitos individuais (aditivo e dominante) e os efeitos individuais e de interação entre os nove SNPs para comparação por meio do teste de razão de verossimilhança (LRT). Foi observado efeito epistático aditivo x dominante (P < 0,01) somente entre os marcadores ARS-BFGL-NGS-71749 e ARS-BFGL-NGS-34135, ambos situados no cromossomo 14 bovino (BTA14), para as características teor de ácidos graxos saturados e teor de ácido palmítico (C16:0). Para o melhor entendimento da sua interação biológica existe ainda a necessidade de maior conhecimento sobre as rotas metabólicas em que tais genes identificados estão situados estes marcadores. Mais estudos nestas regiões cromossômicas possibilitarão ampliar o conhecimento sobre os genes e suas interações. / Bovine milk is an essential food in human nutrition, especially for younger people for being as important source of proteins, minerals and vitamins. The growing for quantity and quality for milk in recent years has stimulated numerous studies, especially regarding the fatty acid profile by differ substantially of human requirements. Nutrition and animal breeding are the main factors which would enhance change the nutritional quality of cow milk. For a long time the changes achieved by animal breeding were obtained purely based on quantitative genetics, but from 90 the interest in this area came to be divided with molecular genetics. The advances made in the molecular biology obtained through the mapping and sequencing of the genomes of different species and genome-wide association studies is helping to explain more about the genetic background of quantitative traits. Currently, most of genome-wide association studies were designed to estimate additive genetic effects only at a single locus, excluding additional effects. However, part of the observed variation in these phenotypes result from the interaction between genes or loci, thus, such studies are also important for the understanding of the origin of biological variation of these traits, such as their metabolic and biochemical pathways. The aims of this study were: I) Associate single nucleotide polymorphisms (SNPs) with production traits and fatty acid profile in milk; and II) Identify interactions between SNPs associated with these traits in milk from Holstein cows. Phenotype adjusted of 760 cows for the fixed effects of contemporary group and covariates days in milk, age at measurement and milk yield, and information from 6,553 SNPs were considered for performing the association using single SNP regression method. A total of nine SNPs were associated with one or more traits related to the fat percentage and fatty acid profile. Some of these associations showed the presence of epistasis and/or pleiotropy between these regions. To identify interaction between SNPs, models that consider the individual effects (additive and dominance); and individual and interaction effects between the nine SNPs for comparison using the likelihood ratio test (LRT). Additive x dominance epistatic effect (P < 0.01) was observed only between markers ARS-BFGL-NGS- 71749 and ARS-BFGL-NGS-34135, both located on chromosome 14 (Bos taurus autossome, BTA14), for fat percentage (FP), saturated fatty acids (SFA) and palmitic acid (C16:0). For a better understanding of its biological interaction there is a need for greater knowledge of the metabolic pathways in with these genes identified these markers are located. More studies will enable these chromosomal regions expand knowledge about genes and their interactions.

O objetivo dessa pesquisa foi avaliar os parâmetros genéticos e correlações das medidas de ingestão e eficiência alimentar com desempenho e características de carcaça, além de realizar o estudo de associação de amplo genoma (GWAS) para ingestão de matéria seca (IMS), consumo alimentar residual (CAR) e ganho de peso (GMD) em bovinos da raça Nelore (Bos indicus). Foram utilizados dados de 1.058 animais com fenótipo para IMS, taxa de conversão alimentar (CA), CAR, ganho de peso residual (GPR), consumo e ganho residuais (CGR), ganho de peso diário (GMD) e características de carcaça. Os parâmetros genéticos da IMS, CA, CAR, GPR e CGR e suas as correlações com GMD e características de carcaça foram estimados utilizando abordagem bayesiana. Quatro marcadores do tipo SNP (Single Nucleotide Polymorphism), localizados em genes relacionados ao controle de apetite (NPY e PDE3B) e transporte iônico (TRPM3 e ITPR1), foram associados com o desempenho, ingestão e medidas de eficiência alimentar. Adicionalmente, os animais foram genotipados em dois chips distintos (Illumina Bovine HD e Illumina BovineSNP50) e as informações genotípicas foram combinadas por meio de estudo de imputação. As centenas de milhares de SNPs foram utilizadas para o GWAS da IMS, CAR e GMD pelo teste de associação GRAMMAR-Gamma. A herdabilidade da IMS, CAR e CGR foi de 0,40, 0,38 e 0,54, respectivamente. Não foram encontradas associações nos SNPs localizados nos genes TRPM3, NPY e ITPR1, no entanto, o SNP no gene PDE3B foi associados significativamente (P&le;0,05) com IMS, CAR e CGR. Os SNPs mais associados com a IMS e CAR, no GWAS, estão localizados nos cromossomos 4, 8, 14 e 21 em regiões genômicas relacionadas com transporte iônico e regulação da ingestão. O GWAS para o GMD apontou os cromossomos 3, 6 e 10 como os que continham os marcadores com maior associação. A ingestão e eficiência alimentar são passíveis de seleção genética, principalmente o CGR. O gene PDE3B deve ser melhor estudado pois, aparenta ter relação com esses fenótipos. Por fim, esse trabalho apontou regiões genômicas, por meio de associação de amplo genoma, relacionadas com a IMS, CAR e GMD, acredita-se ser o primeiro estudo desse tipo para esses fenótipos em animais da raça Nelore. / The aim of this study was to evaluate the genetic parameters and correlations of intake and feed efficiency with performance and carcass traits, and perform the genome-wide association study (GWAS) for dry matter intake (DMI), residual feed intake (RFI) and average daily gain (ADG) in Nelore cattle (Bos indicus). Data from 1,058 animals phenotyped for DMI, feed conversion ratio (FCR), RFI, residual body weight gain (RG), residual intake and body weight gain (RIG), average daily gain (ADG) and carcass traits were used. Genetic parameters of DMI, FCR, RFI, RG and RIG and their correlations with ADG and carcass traits were estimated using Bayesian approach. Four SNPs (Single Nucleotide Polymorphism), located in genes related to appetite control (NPY and PDE3B) and ion transport (TRPM3 and ITPR1), were associated with performance, intake and feed efficiency traits. Additionally, the animals were genotyped in two different chips (Illumina Bovine HD and Illumina BovineSNP50) and genotypic information were combined by imputation. The hundreds of thousands of SNPs were used for GWAS of DMI, RFI and ADG by GRAMMAR-Gamma association test. The heritability of DMI, RFI and RIG was 0.40, 0.38 and 0.54, respectively. No associations were found in SNPs in genes TRPM3, NPY and ITPR1, however, the SNP in PDE3B gene was significantly associated (P&le;0.05) with DMI, RFI and RIG. The SNPs most associated with DMI and RFI, in GWAS, are located on chromosomes 4, 8, 14 and 21 in genomic regions associated with ion transport and appetite regulation. The GWAS pointed chromosomes 3, 6 and 10 as those containing more associated markers for ADG. Feed intake and feed efficiency are amenable to genetic selection, especially the RIG. The PDE3B gene should be further studied thus appears to be related to these phenotypes. Finally, this work shows genomic regions by genome-wide association, related to DMI, RFI and ADG, believed to be the first study of its kind for these phenotypes in Nellore cattle.

O gene DMRT3 tem sido descrito como o principal gene a atuar na determinação da marcha em diversas raças equinas. O alelo A do SNP 23:g.22999655C&gt;A do DMRT3 foi apontado como responsável por essa característica. Na raça brasileira Mangalarga Marchador, a qual apresenta dois padrões de marcha com características bem definidas, os genótipos AA e CA vem sendo associados à marcha picada e o genótipo CC à marcha batida. O objetivo geral do presente prospectar regiões genômicas associadas às marchas batida e picada em equinos Mangalarga Marchador. Foram utilizados 1.230 dados fenotípicos sobre o tipo de andamento (marcha batida N = 1.006; marcha picada N = 227) e, considerando a totalidade da genealogia conhecida para cada animal, 3172 animais no pedigree. Primeiramente foram testadas estratégias de modelagem para esta característica a fim de determinar os efeitos a serem considerados no modelo, bem como a melhor forma de inclusão (efeito fixo ou aleatório). Posteriormente, foi estudada a relação entre as frequências alélicas e genotípicas do gene DMRT3 com os padrões de parentesco e endogamia de acordo com cada tipo de marcha. Um estudo de associação genômica ampla em passo único (considerando informações de animais genotipados e não-genotipados simultaneamente) foi conduzido para verificar regiões genômicas, polimorfismos de nucleotídeo único e genes relacionados com a determinação do tipo de marcha em cavalos Mangalarga Marchador. Vinte e dois polimorfismos de nucleotídeo único localizados nos cromossomos 4(N = 5), 6 (2), 16 (1), 23 (11), 26 (1) e 29 (2), foram responsáveis por 42,43% da variância genética aditiva. Foram associados ao tipo de marcha 69 genes, mas cerca de 39 não estavam anotados em equinos. Foi conduzido um blast a fim de recuperar a função mais provável destes genes. Foram encontradas oito vias metabólicas associadas ao tipo de marcha. Os principais genes envolvidos estavam relacionados à percepção de estímulos externos, metabolismo energético-oxidativo, sistema imune e aprendizado e ritmo da locomoção. Não foi possível identificar a(s) variante(s) causal(ais) do tipo de marcha, contudo este estudo foi o primeiro e verificar que a possível determinação genética do tipo de marcha em cavalos Mangalarga Marchador passa por diferenças em níveis metabólicos que garantem a adaptação dos animais ao tipo de andamento. / The DMRT3 gene has been described as the main gene to act in gait determination in several equine breeds. The allele A of the SNP 23:g.22999655C&gt;A of DMRT3 gene was identified as responsible for this trait. In the Brazilian Mangalarga Marchador breed, which presents two gait patterns with characteristics well defined, the AA and CA genotypes have been associated with marcha picada gait and CC genotype with marcha batida gait. The general aim of this study was to prospect genomic regions associated with marcha batida and marcha picada gaits in Mangalarga Marchador equines. 1,230 phenotypic data were used on the type of gait (marcha batida N = 1.006; marcha picada N = 227) and, considering the totality of known genealogy for each animal, 3,172 animals in the pedigree. Firstly, modelling strategies were tested for this trait in order to determine the effects to be considered in the model, as well as the best form of inclusion (fixed or random effect). Based on the best modelling strategy to be used, the relationship between the allelic and genotypic frequencies of the DMRT3 gene with kinship and inbreeding patterns was studied according to each type of gait. A single-step wide genomic association study (considering information from both genotyped and non-genotyped animals simultaneously) was conducted to verify genomic regions, single nucleotide polymorphisms and genes related to determination of gait type in Mangalarga Marchador horses. Twenty two single nucleotide polymorphisms located on chromosomes 4 (N = 5), 6 (2), 16 (1), 23 (11), 26 (1) and 29 (2) were responsible for 42.43% of the additive genetic variance. 69 genes were associated with gait type, but about 39 were not annotated in horses. A blast was conducted in order to recover the most likely function of these genes. Eight metabolic pathways were found associated with gait type and the main genes involved were related to the perception of external stimuli, energy-oxidative metabolism, immune system and learning and rhythm of locomotion. It was not possible to identify the causal variant(s) of the type of gait; however, this study was the first and to verify that the possible genetic determination of gait type in Mangalarga Marchador horses goes through differences in the metabolic levels that guarantee the adaptation of the animals to the type of gait.

Le potentiel de génération thrombine (TGP en anglais) est une nouvelle mesure qui permet de quantifier in vitro l'activité globale de la thrombine reflétant bien les mécanismes in vivo de la coagulation. Ce méthode de dosage est sensible aux déficits de facteurs de coagulation, à la prise d'anti-coagulants et à de nombreux troubles de la coagulation. Au moment où j'ai débuté ma thèse, seuls deux polymorphismes génétiques, tous les deux situés dans le gène F2 codant pour la prothrombine, étaient connus pour influencer la variabilité plasmatique du TGP. Mon projet de thèse avait pour objectifs d'identifier de nouveaux facteurs génétiques, mais également épigénétiques, pouvant influencer les taux plasmatiques de TGP. Dans une première partie, j'ai mené la toute première étude d'association génome-entier (GWAS pour Genome Wide Association Study en anglais) sur 3 biomarqueurs (temps de latence, quantité totale de thrombine produite et niveau maximal de thrombine produite) du TGP dans deux études françaises rassemblant 1267 sujets et j'ai répliqué les résultats les plus significatifs dans deux autres études françaises indépendantes de 1344 sujets. Cette stratégie a permis de mettre en évidence qu'un polymorphisme génétique du gène ORM1 était associé de manière robuste au temps de latence, biomarqueur caractérisant le temps nécessaire pour initier la coagulation après induction. Dans la seconde partie de ma thèse, en suivant une stratégie similaire mais cette fois-ci en étudiant non plus des polymorphismes génétiques mais des marques de méthylation d'ADN, j'ai recherché si des niveaux de méthylation de site CpG, mesurés à partir d'ADN sanguin et couvrant l'ensemble du génome, pouvaient être associés à la variabilité des 3 mêmes biomarqueurs de TGP. Malheureusement, à partir de deux échantillons mis à ma disposition et rassemblant 425 sujets, je n'ai pas pu mettre en évidence d'association robuste entre des marques de méthylation sanguine et la génération trombine. / Thrombin Generation Potential (TGP) is a promising in vitro measurement that allows quantifying thrombin activity, in a close way to what happens in vivo. It is sensitive to coagulation factors deficiencies, anticoagulant proteins and is associated to thrombotic disorders. There exists two polymorphisms located in the F2 (prothrombin) gene known to influence TGP levels, and altogether they explain 11.3% of the TGP inter-individual variability. With the aims of identifying novel genetic and epigenetic factors that influence TGP variability, I have performed two different studies in the present work. First, I conducted the first genome-wide association study for the three TGP biomarkers (ETP, Peak and Lagtime) using imputation data from two French studies. The most significant single nucleotide polymorphisms (SNPs) were then replicated in two independent French studies. This analysis lead to the discovery of ORM1 as a new gene participating to the control of TGP. Second, I followed a similar strategy using this time whole blood DNA methylation levels at CpG sites to identify DNA methylation marks involved in TGP variability. I analyzed the association between methylation-wide patterns from a French study and a French-Canadian families measured for TGP. Unfortunately, I did not identify robust associations between whole DNA methylation levels and thrombin generation.