Canalization, the stability of phenotypes in the presence of a specific genotype and environment is important for trait selection and understanding gene expression pathways. The goal of this study was to investigate mechanisms involved in the loss of canalization at the maize r1 locus. A specific lightly colored R-sc derivative, R-sc:86-17(L), was mutagenized with ethyl methyl sulfonate (EMS), and the progeny screened for restoration of dark seed color and canalization. Putative revertants/mutants were mapped relative to the r1 locus, tested for complementation to each other to identify a minimum number of genes involved in canalization, tested for cis silencing of the r1 gene Lc, analyzed to determine the genetic basis for loss of canalization and r1 gene expression, analyzed for changes in molecular structure at the r1 locus, and analyzed for changes in cytosine methylation patterns. At least two complementation groups were found to be responsible for increased seed color, one at or near r1 and one elsewhere in the genome. Phenotypic analysis revealed highly variable kernel pigmentation, possible incomplete penetrance and a novel class termed Nc enhancers. Nc enhancers were characterized by light to heavy mottling. Southern blot hybridization analysis demonstrated that 19 out of 20 putative EMS-induced mutants were unchanged in the molecular structure of genes Sc‖nc1, Nc2, and Nc3. One mutant resulted in the loss of one of the Nc genes. The loss of Lc expression was found to be rare in R-sc:86-17(L) Lc, and restoration of seed color in R-sc:86-17(L)/P heterozygotes was found to have resulted from recombination between Sc‖nc1 and P with loss of Nc2 and Nc3. Analysis of cytosine methylation patterns revealed low levels of methylation in the 5’ region of Sc and high levels in the 5’ region of the Nc genes and in the 3’ ends of Sc‖nc1, Nc2 and Nc3 in the homozygous putative EMS-induced mutants as previously found for R-sc:86-17(L). Four out of the twenty EMS-induced mutants tested displayed altered methylation patterns, with changes at both the 5’ and 3’ regions of the r1 genes. These findings suggest that the genes or elements involved in the canalization and/or restoration of seed color of R-sc are located at the r1 locus, at the 3’ end of the Sc gene and that at least one additional site in the genome is involved in seed color expression of r1.

Ce travail de recherche a été mené dans l’équipe MODGEN du département GenPhySE, 1388 UMR de l’INRA. Il avait pour perspectives de modéliser les variabilités de la taille de portée des ovins pour une meilleure évaluation génétique et de proposer des modèles d’analyse adoptés à ce type de caractère.Les objectifs de l’étude étaient d'identifier, et d'analyser les facteurs environnementaux et génétiques des variabilités des tailles de portée des brebis et de proposer une méthode adaptée pour canaliser les caractères discrets qui prenne en compte conjointement des objectifs sur la moyenne et sur la variabilité du caractère. La thèse s’appuyait sur des données issues de la chaîne nationale de contrôle de performances en relation avec des organismes de sélection ovin qui sont particulièrement intéressés par les résultats de ces recherches (CORAM). Choix des races. Nous avons considéré trois races dont les fréquences de la taille de portée sont plus ou moins élevées. La race Romane connue par sa forte prolificité ainsi des effectifs contrôlés importants a permis des analyses précises. Les deux autres races choisies sont la race Suffolk et la race Rouge de l’Ouest qui ont servi à tester les modèles sur des populations comportant de moins en moins de portées multiples. La taille de portée des ovins considérée comme une variable continue. Au cours de l’étude, nous avons identifié et hiérarchisé les facteurs de variation environnementaux du niveau de prolificité et de de sa variance pour trois races ovines (Suffolk, Rouge de l’Ouest et Romane). Si les facteurs de variation environnementaux du niveau de prolificité étaient déjà bien connus, peu de données concernaient les facteurs de variation environnementaux de la variance. Nous avons également estimé les paramètres génétiques du niveau de prolificité et de sa variance, ils sont cohérents entre eux. Les héritabilités estimées sur la moyenne sont faibles, en cohérences entre les trois races et avec des études antérieures. La non homogénéité des corrélations estimées par le modèle à deux étapes peut s’expliquer par la relation structurelle entre la moyenne et la variance sur l’échelle observée qui n'est pas prise en compte dans ce type d'analyse. A l'échelle d'une race, cette relation dépend de l’écart spécifique entre les seuils; elle est différente de la relation moyenne – variance observée pour l’ensemble des races et qui est calculée sur des données brutes. La taille de portée des ovins considérée comme une variable discrète Nous avons montré l’existence d’une variabilité d’origine génétique dans les trois races étudiées et nous avons montré que le modèle à seuils homothétique [HTM] s’ajustait mieux aux données et qu'il permettait de calculer conjointement l’effet génétique sur la moyenne et sur la variance. Nous avons aussi montré les limites de ce modèle et nous avons proposé une solution afin de donner un cadre plus général à son utilisation et l’utiliser dans les évaluations génétiques. Nous avons également identifié un critère de sélection pertinent qui permet de sélectionner les individus de même moyenne ayant des variances différentes. Nous avons proposé le modèle à seuils indépendants [ITM] ou le modèle multi caractères avec contrainte, ce modèle considère que chaque réalisation est un caractère. Nous avons montré la faisabilité de ce modèle sur les trois races. Ceci nous a permis de mettre en évidence les problèmes liés à la convergence du programme utilisé, et nous avons proposé d’implémenter ce modèle en modifiant le programme TM. / This research work was conducted in the MODGEN team of GenPhySE department, 1388 UMR of INRA. The aim of the study was to model the range size variability of sheep for a better genetic evaluation and to propose models of analysis adopted for this type of character. The objectives of the study were to identify, and analyze the environmental and genetic factors of ewe range size variability and to propose a suitable method for channeling discrete traits that takes into account both mean and character variability objectives. The thesis was based on data from the national chain of performance control in relation to sheep breeding organizations that are particularly interested in the results of this research (CORAM). Choice of breeds We have considered three breeds whose litter size frequencies are higher or lower. The Romance breed known for its high prolificity and important controlled numbers allowed precise analyzes. The other two breeds selected are the Suffolk breed and the Western Red breed, which were used to test the models on populations with fewer and fewer litters. Sheep litter size considered a continuous variable During the study, we identified and prioritized the environmental variation factors of the level of prolificacy and its variance for three sheep breeds (Suffolk, Western Red, and Romane). ). While the environmental variation factors of the level of prolificacy were already well known, there was little data on the environmental variance factors of the variance. We have also estimated the genetic parameters of the level of prolificacy and its variance, they are coherent with each other. Heritabilities estimated on the mean are low, consistent between the three races and with previous studies. The non-homogeneity of the correlations estimated by the two-stage model can be explained by the structural relationship between the mean and the variance on the observed scale that is not taken into account in this type of analysis. At the scale of a race, this relation depends on the specific difference between the thresholds; it is different from the mean - variance relationship observed for all breeds and calculated on raw data. Sheep litter size considered a discrete variable We have shown the existence of genetic variability in the three races studied and we have shown that the homothetic threshold model [HTM] fits better with the data and that it allowed to calculate jointly the genetic effect on the mean and on the variance. We have also shown the limitations of this model and have proposed a solution to provide a more general framework for its use and use it in genetic evaluations. We have also identified a relevant selection criterion that makes it possible to select individuals of the same average with different variances. We proposed the model with independent thresholds [ITM] or the multi-character model with constraint, this model considers that each realization is a character. We have shown the feasibility of this model on the three races. This enabled us to highlight the problems related to the convergence of the program used, and we proposed to implement this model by modifying the TM program.

Environmental variation, an important source of phenotypic variation, can influence phenotypes, fitness and even rates of evolution. My thesis explores the effects of evolutionary change on the response to different types of environmental variation. In the first study, I examined the evolution of environmental canalization in adaptively diverged populations of Drosophila melanogaster that vary in degree of genetic canalization. I use these populations to test the congruence hypothesis which predicts that genetic canalization is a co-product of the evolution of environmental canalization and thus should be correlated. My results show that, despite adaptive evolutionary changes in morphology and genetic canalization, these populations have similar degree of variability due to environmental variation. In the second study, I explore how both variation in temperature and food quality during development influence patterns of sexual dimorphism in wing morphology in adaptively diverged populations of Drosophila melanogaster. I compare the relative contributions of adaptation, food availability and temperature on sexual size and shape dimorphism of the Drosophila wing. In particular, I focus on how these factors influence size-shape allometry both in general and in a sex-specific manner. My results show that despite the large adaptive divergence and a strong influence of environmental manipulation on wing size and shape, sex-specific patterns of condition dependence remain relatively consistent between the two populations. In the third study, I explore the evolutionary patterns of condition-dependent sexual size dimorphism among 27 different species from the melanogaster species group with varying degrees of sexual size dimorphism. Using food availability manipulations during development, I examine how sexual size dimorphism changes in response to condition at both the intra-specific and the inter-specific level. The results of this study suggest that, although we see a correlation between sexual size dimorphism and condition dependence among traits within most species, sexual dimorphism and condition dependence do not seem to have a correlated evolution among species of the melanogaster species group. / Dissertation / Doctor of Philosophy (PhD)

No description available.

Em sistemas biológicos, o estudo da estabilidade das redes de regulação gênica é visto como uma contribuição importante que a Matemática pode proporcionar a pesquisas sobre câncer e outras doenças genéticas. Neste trabalho, utilizamos o conceito de ``canalização\'\' como sinônimo de estabilidade em uma rede biológica. Como as características de uma rede de regulação canalizada ainda são superficialmente compreendidas, estudamos esse conceito sob o ponto de vista computacional: propomos um modelo matemático simplificado para descrever o fenômeno e realizamos algumas análises sobre o mesmo. Mais especificamente, a estabilidade da maior bacia de atração das redes Booleanas - um clássico paradigma para a modelagem de redes de regulação - é analisada. Os resultados indicam que a estabilidade da maior bacia de atração está relacionada com dados biológicos sobre o crescimento de colônias de leveduras e que considerações sobre a interação entre as funções Booleanas e a topologia da rede devem ser realizadas conjuntamente na análise de redes estáveis. / In biological systems, the study of gene regulatory networks stability is seen as an important contribution that Mathematics can make to cancer research and that of other genetic diseases. In this work, we consider the concept of ``canalization\'\' as a consequence of stability in gene regulatory networks. The characteristics of canalized regulatory networks are superficially understood. Hence, we study the canalization concept under a computational framework: a simplified model is proposed to describe the phenomenon using Boolean Networks - a classical paradigm to modeling regulatory networks. Specifically, the stability of the largest basin of attraction in gene regulatory networks is analyzed. Our results indicate that the stability of the largest basin of attraction is related to biological data on growth of yeast colonies, and that thoughts about the interaction between Boolean functions and network topologies must be given in the analysis of stable networks.

Em sistemas biológicos, o estudo da estabilidade das redes de regulação gênica é visto como uma contribuição importante que a Matemática pode proporcionar a pesquisas sobre câncer e outras doenças genéticas. Neste trabalho, utilizamos o conceito de ``canalização\'\' como sinônimo de estabilidade em uma rede biológica. Como as características de uma rede de regulação canalizada ainda são superficialmente compreendidas, estudamos esse conceito sob o ponto de vista computacional: propomos um modelo matemático simplificado para descrever o fenômeno e realizamos algumas análises sobre o mesmo. Mais especificamente, a estabilidade da maior bacia de atração das redes Booleanas - um clássico paradigma para a modelagem de redes de regulação - é analisada. Os resultados indicam que a estabilidade da maior bacia de atração está relacionada com dados biológicos sobre o crescimento de colônias de leveduras e que considerações sobre a interação entre as funções Booleanas e a topologia da rede devem ser realizadas conjuntamente na análise de redes estáveis. / In biological systems, the study of gene regulatory networks stability is seen as an important contribution that Mathematics can make to cancer research and that of other genetic diseases. In this work, we consider the concept of ``canalization\'\' as a consequence of stability in gene regulatory networks. The characteristics of canalized regulatory networks are superficially understood. Hence, we study the canalization concept under a computational framework: a simplified model is proposed to describe the phenomenon using Boolean Networks - a classical paradigm to modeling regulatory networks. Specifically, the stability of the largest basin of attraction in gene regulatory networks is analyzed. Our results indicate that the stability of the largest basin of attraction is related to biological data on growth of yeast colonies, and that thoughts about the interaction between Boolean functions and network topologies must be given in the analysis of stable networks.

Vnitřnosti města: pod zemí. Školní interpretace ABSTRACT My diploma thesis deals with subterranean spaces, the insides of a town. It focuses on selected types of a town underground which are described. These types include cellars (common house cellars as well as that of public historical places), a canalization system, metro, catacombs, adits and military underground spaces. I concentrate on the subject through contexts of phenomenology, history, archaeology, literature, and visual arts. The aim of the thesis is mapping subterranean places, trying to find their aspects and using them for a didactic transformation. I try to conceive that by studying the literature as well as by a practical research: I visit several underground places of a different kind. The selected places consist of the personage cellar in Bezděz, the Cemetry church of all Saints in Sedlec - Ossuary, the Corpus Christi Chapel in Kutná Hora, the Old Wastewater Treatment Plant in Bubeneč and the construction tunnel of metro line D in Pankrác. The didactic part transforms basic findings into a number of lessons. These are carried out in a school and out- of-school background. The aim of the didactic project is introducing the topic in its basic forms (mainly cellar, canalization and metro) to the pupils. Hopefully, this can lead them to...

Daphnia is a widespread freshwater zooplankton species, which is both a classic and emerging new model for research in ecological physiology, ecotoxicology and evolutionary biology of adaptation to novel environments. Heat tolerance in Daphnia is known to depend both upon evolutionary history of a genotype and on individuals' acclimation to elevated temperature and to correlate with the level of haemoglobin expression. We demonstrate the existence of north-south gradient of heat tolerance in North American D. pulex, which is not associated with any parallel changes in haemoglobin expression. Geographically distinct clones differ in the way their haemoglobin expression changes due to acclimation to a sub-stressful (28°C) temperature, but these changes are not correlated with the latitude of clones' origin. Likewise, the effect of acclimation to sub-stressful temperature is independent from, and cannot be fully explained by, haemoglobin expression changes during acclimation. The degree of oxidative damage to haemoglobin, measured as the ratio of absorbance at 540:576 nm at the acclimation temperature, is a strong predictor of 28°C-acclimated Daphnia survival during an acute heat exposure. The comparison of haemoglobin expression in resistant and tolerant clones acclimated to different temperatures indicates that tolerant clones exhibit canalization of haemoglobin expression, possessing a high level of haemoglobin even at non-stressful temperatures. We discuss the evolutionary biology of adaptation and acclimation to elevated temperatures in an ecologically important component of freshwater ecosystems in the context of global climate change.

Background: Gene expression regulation is one of the fundamental mechanisms of phenotypic plasticity and is expected to respond to selection in conditions favoring phenotypic response. The observation that many organisms increase their stress tolerance after acclimation to moderate levels of stress is an example of plasticity which has been long hypothesized to be based on adaptive changes in gene expression. We report genome-wide patterns of gene expression in two heat-tolerant and two heat-sensitive parthenogenetic clones of the zooplankton crustacean Daphnia pulex exposed for three generations to either optimal (18°C) or substressful (28°C) temperature. Results: A large number of genes responded to temperature and many demonstrated a significant genotype-byenvironment (GxE) interaction. Among genes with a significant GxE there were approximately equally frequent instances of canalization, i.e. stronger plasticity in heat-sensitive than in heat-tolerant clones, and of enhancement of plasticity along the evolutionary vector toward heat tolerance. The strongest response observed is the across-the-board down-regulation of a variety of genes occurring in heat-tolerant, but not in heat-sensitive clones. This response is particularly obvious among genes involved in core metabolic pathways and those responsible for transcription, translation and DNA repair. Conclusions: The observed down-regulation of metabolism, consistent with previous findings in yeast and Drosophila, may reflect a general compensatory stress response. The associated down-regulation of DNA repair pathways potentially creates a trade-off between short-term benefits of survival at high temperature and long-term costs of accelerated mutation accumulation.

J'étudie la formation du système vasculaire des feuilles des plantes à l'aide de modèles mathématiques. L'hypothèse de canalisation d'une phytohormone, l'auxin, stipule que l'auto activation de son transport entre les cellules crée des chemins préférentiels qui se différencieront plus tard en système vasculaire. J'entreprends une analyse numérique de modèles de canalisation sur une grande matrice et parviens à créer des motifs branchés dans lesquels circule l'auxin. Une analyse de stabilité d'un modèle simplifié nous éclaire sur les raisons de la formation de ces motifs et l'impossibilité de créer un réseau réticulé. La majorité des plantes ayant un système vasculaire réticulé, je modifie le modèle de façon à obtenir ce type de réseau. En ajoutant une variable biologiquement plausible je parviens à créer un réseau réticulé dans lequel l'auxin circule uniformément. Enfin, je discute des relations entre la formation du système vasculaire et de la spirale de phyllotaxie.