Animal traits differ not only in mean, but also in variation around the mean. For instance, one sire’s daughter group may be very homogeneous, while another sire’s daughters are much more heterogeneous in performance. The difference in residual variance can partially be explained by genetic differences. Models for such genetic heterogeneity of environmental variance include genetic effects for the mean and residual variance, and a correlation between the genetic effects for the mean and residual variance to measure how the residual variance might vary with the mean. The aim of this thesis was to develop a method based on double hierarchical generalized linear models for estimating genetic heteroscedasticity, and to apply it on four traits in two domestic animal species; teat count and litter size in pigs, and milk production and somatic cell count in dairy cows. The method developed is fast and has been implemented in software that is widely used in animal breeding, which makes it convenient to use. It is based on an approximation of double hierarchical generalized linear models by normal distributions. When having repeated observations on individuals or genetic groups, the estimates were found to be unbiased. For the traits studied, the estimated heritability values for the mean and the residual variance, and the genetic coefficients of variation, were found in the usual ranges reported. The genetic correlation between mean and residual variance was estimated for the pig traits only, and was found to be favorable for litter size, but unfavorable for teat count.

Na última década, a podridão vermelha das raízes da soja (PVR), ou síndrome da morte súbita, causada pelo fungo Fusarium solani f.sp. glycines, tornou-se uma doença que é motivo de preocupação para os sojicultores, técnicos e pesquisadores nas regiões onde esta doença já foi constatada, sendo a única estratégia de controle viável a utilização de cultivares resistentes. Diante disto, o principal objetivo do presente trabalho foi o estudo do controle genético da reação da soja a PVR por meio de técnicas de genética clássica e molecular. Foi utilizada a geração F2 de um dialelo 5x5, sem os recíprocos, envolvendo cinco cultivares: Forrest, MG/BR-46 (Conquista), IAC-4, FT-Cristalina e FT-Estrela, sendo as duas primeiras mais resistentes a PVR que IAC-4, considerada moderadamente resistente e, as duas últimas, altamente suscetíveis. Além de testes de inoculação com as cultivares, foram conduzidos três experimentos com a geração F2: um em telado (semeadura em julho de 2001) e dois em casa de vegetação (semeadura em setembro de 2001 e julho de 2002), sendo os dois primeiros em blocos ao acaso e o terceiro no delineamento inteiramente casualizado. O patógeno foi inoculado com grãos de sorgo colonizados, colocando-se três grãos no fundo de cada cova, no momento da semeadura, fazendo-se cinco covas por vaso, cada qual constituindo uma parcela com cinco plantas. Em cada experimento, foram avaliadas individualmente 50 plantas de cada genitor e 150 plantas de cada cruzamento F2, entre os 30 e 40 dias após a emergência, utilizando-se uma escala de notas de 1 a 5 para a severidade dos sintomas foliares da PVR. A porcentagem de incidência da doença em cada parcela e um índice de doença também foram calculados. Nas análises de variância com os dados de médias de parcelas, observaram-se diferenças altamente significativas entre os genitores e entre populações F2 para a severidade e índice de doença dos sintomas foliares, na maioria dos casos, embora os genitores resistentes e suscetíveis não foram muito contrastantes. As cultivares Forrest e Conquista comportaram-se sempre como resistentes, e Cristalina e Estrela como suscetíveis, enquanto que IAC-4 apresentou comportamento instável. Pela análise dialélica de Jinks-Hayman reafirmou-se a influência do ambiente sobre o controle da resistência à manifestação dos sintomas foliares da PVR, a qual foi controlada quantitativamente. Nos experimentos de 2001, constatou-se apenas a ação de efeitos gênicos aditivos. Porém, no experimento conduzido em telado, a resistência demonstrou controle por genes recessivos, enquanto que na casa de vegetação, na maior parte, por genes dominantes. No experimento de 2002 constatou-se a presença de efeitos gênicos aditivos e de dominância, predominando o efeito destes últimos. Assim, com base no experimento de 2002, para o grupo de cultivares estudado, os parâmetros genéticos calculados permitiram verificar que: o grau médio de dominância indicou a presença de sobredominância; predominaram genes recessivos no grupo dos genitores; pelo menos três locos ou blocos gênicos que exibiram dominância foram responsáveis pelo controle da resistência a PVR; as herdabilidades estimadas no sentido restrito foram médias (0,33 a 0,62) e, no sentido amplo, altas (0,90 a 0,96), confirmando a presença de dominância; a resistência foi controlada, na maior parte, por genes dominantes e a ordem decrescente de dominância das cultivares foi a seguinte: 'Conquista', 'Cristalina', 'Forrest', 'Estrela' e 'IAC-4'; a exclusão de 'Cristalina', por suspeita de apresentar distribuição de genes correlacionada com 'Conquista' e 'Estrela', melhorou a adequação dos dados de índice de doença ao modelo genético aditivo-dominante de Jinks-Hayman. A utilização de cinco marcadores moleculares microssatélites (Satt163, Satt309, Satt354, Satt371 e Satt570), relatados como ligados a cinco QRLs da PVR, indicou a provável presença de multialelismo nestes locos, o que não invalidou a adequação dos dados ao modelo de Jinks-Hayman. Pela análise de ligação entre 126 indivíduos F2 do cruzamento 'Conquista' x 'Estrela' com os marcadores Satt163 e Satt354, no experimento conduzido em telado, houve ligação fraca (P<0,10) entre estes locos e os respectivos QRLs, havendo tendência dos alelos recessivos serem os responsáveis pelo controle da resistência nestes dois locos, concordando com os resultados da análise dialélica para este experimento. / In order to study the genetic control of soybean resistance to sudden death syndrome (SDS) by classical and molecular genetic techniques a 5x5 diallel with the F2 generation, without the reciprocals, was carried out. The following parents were used: 'Forrest', 'MG/BR-46 (Conquista)', 'IAC-4', 'FT-Cristalina' and 'FT-Estrela'. The first two cultivars are more resistant to SDS than 'IAC-4', that is considered to be moderately resistant to SDS, and the last two cultivars are highly susceptible. Tests of inoculation were done with the cultivars and three experiments with the F2 generation (two in 2001 and one in 2002) were carried out, all of them in greenhouses. The fungus was inoculated by three colonized sorghum grains placed at the bottom of the holes at the planting. It was used five-holes/clay pot, which one was considered a plot with five plants. In each experiment with the F2 generation 50 single plants of each parent and 150 single plants of each F2 population were evaluated between 30 and 40 days after emergency by using a scale (1 to 5) based on foliar severity symptoms. The disease incidence and a disease index also were calculated for each plot. In the ANOVAs with data plot average for severity and disease index highly significant differences were detected among the treatments in almost all cases, although the resistant and the susceptible parents did not differ too much. The parents 'Forrest' and 'Conquista' were always more resistant than the others. 'Cristalina' and 'Estrela' were the most susceptible parents, while 'IAC-4' was unstable. Jinks-Hayman's analysis reaffirmed the environment effect on the genetic control of the resistance to SDS foliar symptoms, which was quantitatively controlled. In the 2001 experiments there was observed only additive genic effects, but in one experiment recessive genes had controlled the resistance, while in the other, in major part, dominant genes had controlled the resistance to SDS. In the 2002 experiment it was showed mainly dominance effects and also some additive genic effects. In the last experiment, for the group of parents used, the genetic parameters indicated that: the average degree of dominance showed the presence of overdominance; there were more recessive than dominants genes in the group of the parents; at least three loci or genic blocks that exhibited dominance were responsible for the genetic control of the resistance to SDS; the heritability in the narrow-sense had middle values (0.33 to 0.62), and in the broad-sense had high values (0.90 to 0.96), reinforcing the presence of dominance effects; the resistance to SDS was controlled, mostly, by dominant genes; the decreasing order of dominance of the parents were: 'Conquista', 'Cristalina', 'Forrest', 'Estrela' and 'IAC-4'; and the exclusion of 'Cristalina' of the diallel for disease index by suspect of gene correlated distribution with 'Conquista' and 'Estrela' improved the fitting of the data to Jinks-Hayman's additive-dominant model. Five microsatellite markers (Satt163, Satt309, Satt354, Satt371 and Satt570), reported as linked to five SDS QRLs, were used and showed the possibility of occurrence of multiallelism in those loci, but this evidence did not invalidate the fitting of the data to Jinks-Hayman's model. The molecular analysis in 126 plants of 'Conquista' x 'Estrela' cross with the markers Satt163 and Satt354, in the first experiment of 2001, showed the tendency of weakly association (P<0,10), between those loci and the QRLs. This analysis showed also tendency that the recessive genes controlled the resistance to SDS in both loci, in according to the results obtained in the diallel analysis for this experiment.

La réponse adaptative de populations de bioagresseurs aux pressions de sélection exercées par les activités agricoles détermine la durabilité des moyens de lutte. Chez le melon, le gène Vat qui confère la résistance à Aphis gossypii étant déployé depuis plus de 10 ans, on craint son contournement. L’enjeu est de proposer des éléments stratégiques aux semenciers sur le risque d’évolution des pucerons vers la virulence, pour développer de nouvelles variétés avec des résistances durables. Dans le cadre de cette thèse, nous avons : i) Estimé la diversité génétique disponible dans des populations d’A. gossypii de différentes régions de production de melon. Elle est structurée géographiquement. La grande diversité observée en France aurait en partie pour origine des évènements de reproduction sexuée suggérant un potentiel évolutif élevé d’A. gossypii. ii) Estimé la pression de sélection exercée par différentes combinaisons de résistance (gène Vat et QTL) sur ces populations. Les densités de population sont plus faibles sur les plantes Vat que sur les plantes non Vat et la structure génétique des populations est modifiée dans certaines régions de production quand le gène Vat est présent. Les clones se multipliant sur les plantes Vat ont une forte fitness et le risque de leurs extensions est grand. Aucun effet de QTL de résistance n’a été mis en évidence en plein champ. iii) Caractérisé les clones contournant le gène Vat. Nos résultats suggèrent que l’adaptation des clones s’effectue soit par modification du gène d’avirulence du puceron soit par l’adaptation du puceron aux effecteurs de la résistance. De nouvelles stratégies de gestion de la résistance Vat sont proposées. / The adaptive response of pest populations to selection pressures exerted by agricultural activities determines the sustainability of control methods. In melon, the Vat gene that confers resistance to Aphis gossypii has been deployed for over 10 years, so there are fears it will be overcome. The challenge is to provide strategic elements to plant breeders, concerning the risk of development of virulent aphids, in order to develop new varieties with durable resistances. In the context of this PhD, we have : i) Estimated the available genetic diversity in populations of A. gossypii from different melongrowing areas. The diversity is structured geographically. The great diversity observed in France would have its origine in part from the events of sexual reproduction, suggesting a high evolutionary potential of A. gossypii. ii) Estimated the selection pressure exerted by different resistance combinations (Vat gene and QTLs) on these populations. Population densities are lower on VatR plants than VatS plants and population genetic structure is altered in certain growing areas when the VatR gene is present. The clones multiplying on VatR plants have good fitness and the risk of their spreading is great. No effect of QTLs has been identified in the field. iii) Characterized the clones overcoming the VatR gene. Our results suggest that the adaptation of clones made either by alteration of the avirulence gene of aphids or by adaptation of aphids toresistance effectors. New strategies for Vat resistance management are proposed.

Force est de constater que les dépérissements forestiers augmentent. Ces observations vont de pairs avec l’accroissement des événements climatiques extrêmes. Aussi dans ce contexte, il est nécessaire d’identifier denouveaux caractères de resistance à la sécheresse. La résistance à la cavitation est actuellement le meilleurmarqueur de la survie d’une espèce à la sécheresse.Cette thèse avait deux objectifs : (i) comprendre le mécanisme de propagation de la cavitation dans le xylèmechez les gymnospermes. (ii) Quantifier la variation phénotypique in situ de ce caractère chez Pinus pinaster ainsique (iii) quantifier la variation génétique, et sa plasticité phénotypique.La démarché a été la suivante (i) une étude interspécifique de la résistance à la cavitation a été couplé à desmesures micro-anatomiques. (ii) Pour le volet intraspécifique, nous avons phénotypé 6 populations dans deuxtest de populations-descendances, ainsi qu’en population naturelles in situ.La propagation de l’embolie chez les Pinaceae et les ex-Taxodiaceae pourrait être due au passage du germe d’air(rupture capillaire) à travers des nanopores dans le torus. En effet, la pression de rupture d’un ménisque air-sèveest corrélée à l’entrée de l’air dans le xylème (P12). Alors que la variation interspécifique est grande, la résistanceà la cavitation varie faiblement au sein d’une espèce. Ainsi les populations provenant de climat contrasté neprésentent pas ou peu de différence génétique (en test de provenance) ou en populations naturelles in situ. Cecaractère présente une plasticité phénotypique mais faible comparée à celle de la croissance en hauteur parexemple. La comparaison entre la variation génétique entre populations et la variation des marqueurs neutresentre ces mêmes populations montrent que la variation de ce caractère semble réduite par l’architecturegénétique sous-jacente. La resistance à la cavitation est vraisemblablement un trait canalisé. / Several review reported global forest die-back that are caused, directly or indirectly, by extreme climatic events(like heat waves or prolonged drought). In this context, there is an urgent need to identify new traits to tracedrought tolerance. Resistance to cavitation is one of the best proxy for survival during extreme drought.The aim of this work was (i) to understand how spreads cavitation in the vascular pathway of gymnosperms (ii)to quantify the phenotypic variation of resistance to cavitation for Pinus pinaster species, (iii) to determine theamount of the genetic variation and phenotypic plasticity available for this trait.A micro-anatomy study was coupled to measurement of resistance to cavitation for various species to foundwhere air-seeding occurs in the bordered pit. To quantify the variability of resistance to cavitation, wephenotyped 506 genotypes using to replicated provenance-progeny trials and on natural in situ populations.The spread of embolism for Pinaceae and ex-Taxodiaceae could be due to minute pore in tori, which are remainsof secondary plasmodesmata. We found that the pressure needed to break a water/air meniscus in these minutepores is correlated with the xylem air entry (P12). Despite the great variability of resistance to cavitation betweenspecies, we found low variability within species. Most of the variability is within population, rather than betweenpopulations. The phenotypic plasticity of resistance to cavitation is low compare to growth traits. Comparisonbetween QST and FST shows that populations exhibit less variation compare to what it is expected under geneticdrift. The variation of resistance to cavitation seems to be narrowed by the genetic architecture, which is the signof canalisation.

<p>Plant height is an important agronomic trait and a major target for crop improvement. Owing to the ease of detection and measurement of plant stature, as well as its high heritability, several height-related mutants have been reported in maize. The genes underlying a few of those mutants have also been identified, with a majority of them related to the biosynthesis or signaling of two key phytohormones - gibberellins (GAs) and brassinosteroids (BRs). However, most other maize dwarfing mutants, and especially those that result from gain-of-function mutations, remain uncharacterized. The present study was undertaken to characterize a novel dominant dwarfing mutant, named <i>D13</i>. This mutant appeared in the M1 population of the inbred B73 that was generated by mutagenesis with ethyl methanesulfonate (EMS). Like most other maize dwarfing mutants, the reduction in <i>D13</i> height was largely due to the compression of the internodes. However, unlike the GA or BR mutants, <i>D13</i> had no defects in the female or male inflorescences. Further, in contrast to the GA and BR mutants, the mesocotyl elongation during etiolation was not impacted in <i>D13</i>. <i>D13</i> seedlings developed red coloration in two to three lowermost leaves. In addition, <i>D13</i> also showed enhanced tillering when the phenotype was very severe. The size of the shoot apical meristem of <i>D13</i> was reduced slightly, and significant aberrations in the structure of vascular bundles in the mutant were observed. All anatomical and phenotypic features of <i>D13</i> were highly exaggerated in homozygous state, indicating the partially dominant nature of the <i>D13</i> mutation. Interestingly, the heterozygous mutants showed remarkable variation in their phenotype, which was maintained across generations. Moreover, the <i>D13</i> phenotype was found to be sensitive to the genetic background, being completely suppressed in Mo17, Oh7B, enhanced in CML322, P39 and changed to different degrees in others. To identify the genetic defect responsible for the <i>D13</i> mutant phenotype, a map-based cloning approach was used, which identified a single base-pair change from G to A (G2976A) in the coding region of a glutamate receptor gene (Zm00001d015007). The G2976A missense mutation resulted in the replacement of alanine with threonine at the location 670. The replaced alanine is highly conserved in glutamate receptors across all domains of life from cyanobacteria to plants to mammals, suggesting a causal relationship between the G2976A substitution and the <i>D13</i> phenotype. To validate this relationship, a targeted EMS-based mutagenesis approach was used to knock-out (inactivate) the <i>D13</i> mutant allele. A suppressor mutant was found in which the <i>D13</i> mutant phenotype reverted to the normal tall phenotype. The sequence of the revertant allele, designated <i>D13</i>*, revealed that the original <i>D13</i> mutant allele underwent a second G to A mutation (G1520A) to change glycine into aspartic acid at position 473. This intragenic second-site mutation in the <i>D13</i> allele suppressed the function of the <i>D13</i> allele, thereby preventing it from interfering with the function of the wild type allele. To further unveil the genes and underlying mechanisms that enable the <i>D13</i> mutant to confer a dwarf phenotype, transcriptomic and metabolomic analyses of <i>D13</i> mutants were conducted and compared to the wild type sibs. While the omics analysis confirmed that stress responses were upregulated and genes related to shoot system development were downregulated in the mutant, the data did not allow us to pinpoint the underlying mechanisms that connect the <i>D13</i> mutation with its dwarfing phenotype. Furthermore, it remains unclear whether these stress and shoot system-related changes result in the manifestation of <i>D13</i> phenotype, or the dwarf phenotype due to <i>D13</i> mutation activates the stress-related mechanisms. This is the first study that signifies the importance of a glutamate receptor gene in controlling plant height.</p>

Les traits quantitatifs complexes sont des caractéristiques mesurables d’organismes vivants qui résultent de l’interaction entre plusieurs gènes et facteurs environnementaux. Les locus génétiques liés à un caractère complexe sont appelés «locus de traits quantitatifs » (QTL). Récemment, en considérant les niveaux d’expression tissulaire de milliers de gènes comme des traits quantitatifs, il est devenu possible de détecter des «QTLs d’expression» (eQTL). Alors que ces derniers ont été considérés comme des phénotypes intermédiaires permettant de mieux comprendre l’architecture biologique des traits complexes, la majorité des études visent encore à identifier une mutation causale dans un seul gène. Cette approche ne peut remporter du succès que dans les situations où le gène incriminé a un effet majeur sur le trait complexe, et ne permet donc pas d’élucider les situations où les traits complexes résultent d’interactions entre divers gènes. Cette thèse propose une approche plus globale pour : 1) tenir compte des multiples interactions possibles entre gènes pour la détection de eQTLs et 2) considérer comment des polymorphismes affectant l’expression de plusieurs gènes au sein de groupes de co-expression pourraient contribuer à des caractères quantitatifs complexes. Nos contributions sont les suivantes : Nous avons développé un outil informatique utilisant des méthodes d’analyse multivariées pour détecter des eQTLs et avons montré que cet outil augmente la sensibilité de détection d’une classe particulière de eQTLs. Sur la base d’analyses de données d’expression de gènes dans des tissus de souris recombinantes consanguines, nous avons montré que certains polymorphismes peuvent affecter l’expression de plusieurs gènes au sein de domaines géniques de co-expression. En combinant des études de détection de eQTLs avec des techniques d’analyse de réseaux de co-expression de gènes dans des souches de souris recombinantes consanguines, nous avons montré qu’un locus génétique pouvait être lié à la fois à l’expression de plusieurs gènes au niveau d’un domaine génique de co-expression et à un trait complexe particulier (c.-à-d. la masse du ventricule cardiaque gauche). Au total, nos études nous ont permis de détecter plusieurs mécanismes par lesquels des polymorphismes génétiques peuvent être liés à l’expression de plusieurs gènes, ces derniers pouvant eux-mêmes être liés à des traits quantitatifs complexes. / Complex quantitative traits are measurable characteristics of living organisms resulting from the interaction between multiple genes and environmental factors. Genetic loci associated with complex trait are called "quantitative trait loci" (QTL). Recently, considering the expression levels of thousands of genes as quantitative traits, it has become possible to detect "expression QTLs " (eQTL). These eQTL are considered intermediate phenotypes and are used to better understand the biological architecture of complex traits. However the majority of studies still try to identify a causal mutation in a single gene. This approach can only meet success in situations where the gene incriminate as a major effect on the complex trait, and therefore can not elucidate the situations where complex traits result from interactions between various genes. This thesis proposes a more comprehensive approach to: 1) take into account the possible interactions between multiple genes for the detection of eQTLs and 2) consider how polymorphisms affecting the expression of several genes in a module of co-expression may contribute to quantitative complex traits. Our contributions are as follows: We have developed a tool using multivariate analysis techniques to detect eQTLs, and have shown that this tool increases the sensitivity of detection of a particular class of eQTLs. Based on the data analysis of gene expression in recombinant inbred strains mice tissues, we have shown that some polymorphisms may affect the expression of several genes in domain of co-expression. Combining eQTLs detection studies with network of co-expression genes analysis in recombinant inbred strains mice, we showed that a genetic locus could be linked to both the expression of multiple genes at a domain of gene co-expression and a specific complex trait (i.e. left ventricular mass). Our studies have detected several mechanisms by which genetic polymorphisms may be associated with the expression of several genes, and may themselves be linked to quantitative complex traits.

Causal models have been used in different areas of knowledge in order to comprehend the causal associations between variables. Over the past decades, the amount of studies using these models have been growing a lot, especially those related to biological systems where studying and learning causal relationships among traits are essential for predicting the consequences of interventions in such system. Graph analysis (GA) and structural equation modeling (SEM) are tools used to explore such associations. While GA allows searching causal structures that express qualitatively how variables are causally connected, fitting SEM with a known causal structure allows to infer the magnitude of causal effects. Also SEM can be viewed as multiple regression models in which response variables can be explanatory variables for others. In quantitative genetics studies, SEM aimed to study the direct and indirect genetic effects associated to individuals through information related to them, beyond the observed characteristics, such as the kinship relations. In those studies typically the assumptions of linear relationships among traits are made. However, in some scenarios, nonlinear relationships can be observed, which make unsuitable the mentioned assumptions. To overcome this limitation, this paper proposes to use a mixed effects polynomial structural equation model, second or superior degree, to model those nonlinear relationships. Two studies were developed, a simulation and an application to real data. The first study involved simulation of 50 data sets, with a fully recursive causal structure involving three characteristics in which linear and nonlinear causal relations between them were allowed. The second study involved the analysis of traits related to dairy cows of the Holstein breed. Phenotypic relationships between traits were calving difficulty, gestation length and also the proportion of perionatal death. We compare the model of multiple traits and polynomials structural equations models, under different polynomials degrees in order to assess the benefits of the SEM polynomial of second or higher degree. For some situations the inappropriate assumption of linearity results in poor predictions of the direct, indirect and total of the genetic variances and covariance, either overestimating, underestimating, or even assign opposite signs to covariances. Therefore, we conclude that the inclusion of a polynomial degree increases the SEM expressive power. / Modelos causais têm sido muitos utilizados em estudos em diferentes áreas de conhecimento, a fim de compreender as associações ou relações causais entre variáveis. Durante as últimas décadas, o uso desses modelos têm crescido muito, especialmente estudos relacionados à sistemas biológicos, uma vez que compreender as relações entre características são essenciais para prever quais são as consequências de intervenções em tais sistemas. Análise do grafo (AG) e os modelos de equações estruturais (MEE) são utilizados como ferramentas para explorar essas relações. Enquanto AG nos permite buscar por estruturas causais, que representam qualitativamente como as variáveis são causalmente conectadas, ajustando o MEE com uma estrutura causal conhecida nos permite inferir a magnitude dos efeitos causais. Os MEE também podem ser vistos como modelos de regressão múltipla em que uma variável resposta pode ser vista como explanatória para uma outra característica. Estudos utilizando MEE em genética quantitativa visam estudar os efeitos genéticos diretos e indiretos associados aos indivíduos por meio de informações realcionadas aos indivíduas, além das característcas observadas, como por exemplo o parentesco entre eles. Neste contexto, é tipicamente adotada a suposição que as características observadas são relacionadas linearmente. No entanto, para alguns cenários, relações não lineares são observadas, o que torna as suposições mencionadas inadequadas. Para superar essa limitação, este trabalho propõe o uso de modelos de equações estruturais de efeitos polinomiais mistos, de segundo grau ou seperior, para modelar relações não lineares. Neste trabalho foram desenvolvidos dois estudos, um de simulação e uma aplicação a dados reais. O primeiro estudo envolveu a simulação de 50 conjuntos de dados, com uma estrutura causal completamente recursiva, envolvendo 3 características, em que foram permitidas relações causais lineares e não lineares entre as mesmas. O segundo estudo envolveu a análise de características relacionadas ao gado leiteiro da raça Holandesa, foram utilizadas relações entre os seguintes fenótipos: dificuldade de parto, duração da gestação e a proporção de morte perionatal. Nós comparamos o modelo misto de múltiplas características com os modelos de equações estruturais polinomiais, com diferentes graus polinomiais, a fim de verificar os benefícios do MEE polinomial de segundo grau ou superior. Para algumas situações a suposição inapropriada de linearidade resulta em previsões pobres das variâncias e covariâncias genéticas diretas, indiretas e totais, seja por superestimar, subestimar, ou mesmo atribuir sinais opostos as covariâncias. Portanto, verificamos que a inclusão de um grau de polinômio aumenta o poder de expressão do MEE.

Les traits quantitatifs complexes sont des caractéristiques mesurables d’organismes vivants qui résultent de l’interaction entre plusieurs gènes et facteurs environnementaux. Les locus génétiques liés à un caractère complexe sont appelés «locus de traits quantitatifs » (QTL). Récemment, en considérant les niveaux d’expression tissulaire de milliers de gènes comme des traits quantitatifs, il est devenu possible de détecter des «QTLs d’expression» (eQTL). Alors que ces derniers ont été considérés comme des phénotypes intermédiaires permettant de mieux comprendre l’architecture biologique des traits complexes, la majorité des études visent encore à identifier une mutation causale dans un seul gène. Cette approche ne peut remporter du succès que dans les situations où le gène incriminé a un effet majeur sur le trait complexe, et ne permet donc pas d’élucider les situations où les traits complexes résultent d’interactions entre divers gènes. Cette thèse propose une approche plus globale pour : 1) tenir compte des multiples interactions possibles entre gènes pour la détection de eQTLs et 2) considérer comment des polymorphismes affectant l’expression de plusieurs gènes au sein de groupes de co-expression pourraient contribuer à des caractères quantitatifs complexes. Nos contributions sont les suivantes : Nous avons développé un outil informatique utilisant des méthodes d’analyse multivariées pour détecter des eQTLs et avons montré que cet outil augmente la sensibilité de détection d’une classe particulière de eQTLs. Sur la base d’analyses de données d’expression de gènes dans des tissus de souris recombinantes consanguines, nous avons montré que certains polymorphismes peuvent affecter l’expression de plusieurs gènes au sein de domaines géniques de co-expression. En combinant des études de détection de eQTLs avec des techniques d’analyse de réseaux de co-expression de gènes dans des souches de souris recombinantes consanguines, nous avons montré qu’un locus génétique pouvait être lié à la fois à l’expression de plusieurs gènes au niveau d’un domaine génique de co-expression et à un trait complexe particulier (c.-à-d. la masse du ventricule cardiaque gauche). Au total, nos études nous ont permis de détecter plusieurs mécanismes par lesquels des polymorphismes génétiques peuvent être liés à l’expression de plusieurs gènes, ces derniers pouvant eux-mêmes être liés à des traits quantitatifs complexes. / Complex quantitative traits are measurable characteristics of living organisms resulting from the interaction between multiple genes and environmental factors. Genetic loci associated with complex trait are called "quantitative trait loci" (QTL). Recently, considering the expression levels of thousands of genes as quantitative traits, it has become possible to detect "expression QTLs " (eQTL). These eQTL are considered intermediate phenotypes and are used to better understand the biological architecture of complex traits. However the majority of studies still try to identify a causal mutation in a single gene. This approach can only meet success in situations where the gene incriminate as a major effect on the complex trait, and therefore can not elucidate the situations where complex traits result from interactions between various genes. This thesis proposes a more comprehensive approach to: 1) take into account the possible interactions between multiple genes for the detection of eQTLs and 2) consider how polymorphisms affecting the expression of several genes in a module of co-expression may contribute to quantitative complex traits. Our contributions are as follows: We have developed a tool using multivariate analysis techniques to detect eQTLs, and have shown that this tool increases the sensitivity of detection of a particular class of eQTLs. Based on the data analysis of gene expression in recombinant inbred strains mice tissues, we have shown that some polymorphisms may affect the expression of several genes in domain of co-expression. Combining eQTLs detection studies with network of co-expression genes analysis in recombinant inbred strains mice, we showed that a genetic locus could be linked to both the expression of multiple genes at a domain of gene co-expression and a specific complex trait (i.e. left ventricular mass). Our studies have detected several mechanisms by which genetic polymorphisms may be associated with the expression of several genes, and may themselves be linked to quantitative complex traits.

Causal models have been used in different areas of knowledge in order to comprehend the causal associations between variables. Over the past decades, the amount of studies using these models have been growing a lot, especially those related to biological systems where studying and learning causal relationships among traits are essential for predicting the consequences of interventions in such system. Graph analysis (GA) and structural equation modeling (SEM) are tools used to explore such associations. While GA allows searching causal structures that express qualitatively how variables are causally connected, fitting SEM with a known causal structure allows to infer the magnitude of causal effects. Also SEM can be viewed as multiple regression models in which response variables can be explanatory variables for others. In quantitative genetics studies, SEM aimed to study the direct and indirect genetic effects associated to individuals through information related to them, beyond the observed characteristics, such as the kinship relations. In those studies typically the assumptions of linear relationships among traits are made. However, in some scenarios, nonlinear relationships can be observed, which make unsuitable the mentioned assumptions. To overcome this limitation, this paper proposes to use a mixed effects polynomial structural equation model, second or superior degree, to model those nonlinear relationships. Two studies were developed, a simulation and an application to real data. The first study involved simulation of 50 data sets, with a fully recursive causal structure involving three characteristics in which linear and nonlinear causal relations between them were allowed. The second study involved the analysis of traits related to dairy cows of the Holstein breed. Phenotypic relationships between traits were calving difficulty, gestation length and also the proportion of perionatal death. We compare the model of multiple traits and polynomials structural equations models, under different polynomials degrees in order to assess the benefits of the SEM polynomial of second or higher degree. For some situations the inappropriate assumption of linearity results in poor predictions of the direct, indirect and total of the genetic variances and covariance, either overestimating, underestimating, or even assign opposite signs to covariances. Therefore, we conclude that the inclusion of a polynomial degree increases the SEM expressive power. / Modelos causais têm sido muitos utilizados em estudos em diferentes áreas de conhecimento, a fim de compreender as associações ou relações causais entre variáveis. Durante as últimas décadas, o uso desses modelos têm crescido muito, especialmente estudos relacionados à sistemas biológicos, uma vez que compreender as relações entre características são essenciais para prever quais são as consequências de intervenções em tais sistemas. Análise do grafo (AG) e os modelos de equações estruturais (MEE) são utilizados como ferramentas para explorar essas relações. Enquanto AG nos permite buscar por estruturas causais, que representam qualitativamente como as variáveis são causalmente conectadas, ajustando o MEE com uma estrutura causal conhecida nos permite inferir a magnitude dos efeitos causais. Os MEE também podem ser vistos como modelos de regressão múltipla em que uma variável resposta pode ser vista como explanatória para uma outra característica. Estudos utilizando MEE em genética quantitativa visam estudar os efeitos genéticos diretos e indiretos associados aos indivíduos por meio de informações realcionadas aos indivíduas, além das característcas observadas, como por exemplo o parentesco entre eles. Neste contexto, é tipicamente adotada a suposição que as características observadas são relacionadas linearmente. No entanto, para alguns cenários, relações não lineares são observadas, o que torna as suposições mencionadas inadequadas. Para superar essa limitação, este trabalho propõe o uso de modelos de equações estruturais de efeitos polinomiais mistos, de segundo grau ou seperior, para modelar relações não lineares. Neste trabalho foram desenvolvidos dois estudos, um de simulação e uma aplicação a dados reais. O primeiro estudo envolveu a simulação de 50 conjuntos de dados, com uma estrutura causal completamente recursiva, envolvendo 3 características, em que foram permitidas relações causais lineares e não lineares entre as mesmas. O segundo estudo envolveu a análise de características relacionadas ao gado leiteiro da raça Holandesa, foram utilizadas relações entre os seguintes fenótipos: dificuldade de parto, duração da gestação e a proporção de morte perionatal. Nós comparamos o modelo misto de múltiplas características com os modelos de equações estruturais polinomiais, com diferentes graus polinomiais, a fim de verificar os benefícios do MEE polinomial de segundo grau ou superior. Para algumas situações a suposição inapropriada de linearidade resulta em previsões pobres das variâncias e covariâncias genéticas diretas, indiretas e totais, seja por superestimar, subestimar, ou mesmo atribuir sinais opostos as covariâncias. Portanto, verificamos que a inclusão de um grau de polinômio aumenta o poder de expressão do MEE.

<p>The goal of this thesis is to algorithmically identify candidate modifiers for <i>retinitis pigmentosa</i> (RP) to help improve therapy and predictions for this genetic disorder that may lead to a complete loss of vision. A current research by (Chow et al., 2016) focused on the genetic contributors to RP by trying to recognize a correlation between genetic modifiers and phenotypic variation in female <i>Drosophila melanogaster</i>, or fruit flies. In comparison to the genome-wide association analysis carried out in Chow et al.’s research, this study proposes using a K-Means clustering algorithm on RNA expression data to better understand which genes best exhibit characteristics of the RP degenerative model. Validating this algorithm’s effectiveness in identifying suspected genes takes priority over their classification.</p><p>This study investigates the linear relationship between <i>Drosophila </i>eye size and genetic expression to gather statistically significant, strongly correlated genes from the clusters with abnormally high or low eye sizes. The clustering algorithm is implemented in the R scripting language, and supplemental information details the steps of this computational process. Running the mean eye size and genetic expression data of 18,140 female <i>Drosophila</i> genes and 171 strains through the proposed algorithm in its four variations helped identify 140 suspected candidate modifiers for retinal degeneration. Although none of the top candidate genes found in this study matched Chow’s candidates, they were all statistically significant and strongly correlated, with several showing links to RP. These results may continue to improve as more of the 140 suspected genes are annotated using identical or comparative approaches.</p>