Epidermal appendages are found on every vertebrate this world has to offer. In fish, these are commonly represented by scales. While we have a solid grasp of how scales develop, little is known about the underlying genetic mechanisms behind these phenotypic changes. Using two species of African cichlids (Labeotropheus fuelleborni and Tropheops “red cheek”) with varying scale phenotypes, we sought to examine their F2 hybrid offspring and statistically link the responsible genetic elements to their respective parental phenotypes through Quantitative Loci Trait (QTL) analysis. Scales were removed from six different locations across the midline of each individual. Then, numerous traits on each scale were measured, and these values were used in the QTL analysis. 42 significant QTL were identified, with multiple QTL intervals possessing promising candidate genes. These genes include: fgfr1b, efna5a, TGIF1, eIF6, and col1a1a. Previous studies have implicated these particular genes and gene families to play important roles in scale and placode development. However, they represent the minority of QTL intervals discovered, providing direction for future research towards the other QTL intervals represented by this study.

Pre-harvest contamination of maize grain with aflatoxin is a chronic problem worldwide and particularly in the southeastern U.S. Aflatoxin is a mycotoxin produced by the fungus Aspergillus flavus, an opportunistic ear-rot pathogen of maize (Zea mays). Resistance to aflatoxin accumulation is heritable, and resistant germplasm-lines are available. These lines are derived from “exotic” genetic backgrounds and were released as sources of resistance, not parental inbreds. However, all current sources of resistance are quantitative, which complicates conventional efforts to introgress resistance alleles from unadapted but resistant donor lines to adapted but susceptible recipient lines. Mapping quantitative trait loci (QTL) and their linked markers enables targeted introgression of the desired alleles via marker-assisted selection. Quantitative trait loci were identified in two F2:3 mapping populations, derived from crossing resistant inbreds Mp715 and Mp717 to a common susceptible parent (Va35). The Mp715 x Va35 population was phenotyped for aflatoxin accumulation under artificial inoculation in replicated field trials at Mississippi State (MSU) in 2015 and 2016. The Mp717 x Va35 population was phenotyped at MSU and Lubbock, TX in 2016. Populations were genotyped using simple sequence repeat (SSR) and single nucleotide polymorphism (SNP) markers and linkage maps created in JoinMap4. To locate QTL, linkage maps, genotypes, and phenotypes were analyzed jointly in QTL Cartographer 2.5 using composite interval mapping (CIM) and multiple interval mapping (MIM) procedures. Five QTL with the beneficial allele contributed by Mp715 were identified during CIM in bins 5.01, 6.06, 7.03 10.04 and 10.05. Three QTL with the beneficial allele contributed by Mp717 were identified during CIM in bins 3.07/3.08, 7.02/7.03, and 10.05. In both populations, QTL were identified with the beneficial allele contributed by Va35. Those QTL did not co-locate across populations but four of the six were on chromosome 1. Significant QTL effects from CIM were used as the initial model terms in MIM, where all QTL effects were fit simultaneously and their gene-action and epistatic interactions estimated.

京都大学 / 新制・課程博士 / 博士(農学) / 甲第23618号 / 農博第2481号 / 新制||農||1088(附属図書館) / 学位論文||R4||N5366(農学部図書室) / 京都大学大学院農学研究科地域環境科学専攻 / (主査)教授 日本 典秀, 教授 田中 千尋, 准教授 刑部 正博 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DGAM

El color de la sémola y harina de trigo candeal se debe fundamentalmente al contenido de pigmentos carotenoides presentes (CPC) en los granos y a su degradación por enzimas oxidativas como las lipoxigenasas. Esta tesis tuvo como objetivo mapear QTL asociados al contenido de pigmentos carotenoides y color amarillo en trigo candeal y encontrar marcadores moleculares ligados a los mismos. Para ello se utilizó una población de 93 RILs (líneas recombinantes endocriadas) derivadas de la cruza de progenitores contrastantes (UC1113 x Kofa) para estos caracteres. Se determinó el color amarillo (CIE b*) y el CPC en harina integral sobre la población de RILs, los progenitores y 8 cultivares comerciales nacionales. Estos materiales fueron sembrados en tres localidades (ACA-Cabildo, CEI-Barrow e INTA-Balcarce) durante dos campañas consecutivas (2006/07 y 2007/08). El test de ANOVA mostró diferencias entre las RILs para el CIE b* y el CPC en cada ambiente. No se observaron efectos significativos del año y la localidad sobre el CIE b*, aunque las interacciones dobles y triples fueron altamente significativas (p< 0,01). El CPC fue afectado en forma altamente significativa por el genotipo, mientras que el efecto del año y localidad fueron significativos (p< 0,05). Para este carácter no se observó interacción del genotipo con el año, pero la interacción del genotipo con la localidad fue significativa y las interacciones localidad x año y genotipo x año x localidad fueron altamente significativas. La heredabilidad fue alta en todos los casos (h2<0,94). Se detectó un elevado número de QTL asociados al CIE b* y al CPC, ubicados sobre los cromosomas 1BL, 2AS, 4AL (2), 5AS, 5BL, 6AL (3), 7AS, 7AL, 7BS, 7BL. Si bien se expresaron de forma diferente según los ambientes, en su mayoría se localizaron en posiciones estables, flanqueados por los mismos marcadores moleculares. Este es el primer informe de la existencia de un QTL sobre el brazo corto del cromosoma 7A, asociado al CPC y al CIE b*. La presencia del QTL 7BS, región homeóloga, solo fue informada recientemente. Los análisis mostraron que los principales QTL significativos según el mapeo combinado (4AL, 6AL, 7BL) poseen también un efecto de interacción. El QTL encontrado sobre el cromosoma 6A (6AL.2) explicó la mayor parte de la variación fenotípica para ambos caracteres sobre el promedio de los seis ambientes. Los resultados del mapeo para el QTL 7BL apoyan la hipótesis de la presencia de un gen adicional al Psy-B1 (involucrado en la biosíntesis de carotenoides) afectando el CPC en esta región. Los marcadores flanqueantes a las regiones genómicas encontradas en el análisis combinado, sobre los cromosomas 6AL (Barc113-wmc553), 7AS (BQ170462_176-Barc174), 7BS (Barc23a-Barc72) y 7BL (PsyB-cfa2257), mostraron ser de utilidad para seleccionar ambos caracteres, CPC y CIE b*. El número de ambientes requeridos para realizar el mapeo de QTL es un factor importante a tener en cuenta, ya que los mismos genotipos presentan la expresión de genes/QTL diferentes en los distintos ambientes. Las regiones genómicas de mayor efecto asociadas a CIE b* y CPC fueron coincidentes, lo que demuestra un control genético similar para ambos caracteres. / Durum wheat flour and semolina bright yellow color is mostly due to the carotenoid pigment content (CPC) in wheat grains and its degradation by oxidative enzymes such as lipoxygenases. The aim of this research work was to map QTL associated with carotenoid pigments content and bright yellow color in durum wheat and to find linked molecular markers to be used in marker assisted selection. A mapping population of 93 RILs (recombinant inbred lines) derived from a cross between parents contrasting for the traits (UC1113 x Kofa) was used. CIE b* and the CPC were evaluated in flour of a mapping population, in both parents and in eight national commercial cultivars. The experimental design consisted in these 103 plant materials planted in three locations (ACA-Cabildo, CEI-Barrow and INTA-Balcarce) during two consecutive growing seasons (2006/07 y 2007/08). The differences between RILs in each location according to the ANOVA test were highly significant. The combined ANOVA test taking into account RILs, years and locations revealed differences for both variables. For CIE b* values, significant effects were not observed for years and locations; however, double and triple interactions were highly significant (p< 0.01). The CPC showed highly significant differences for genotype effects while these differences were significant for year and location (p< 0.05). For this trait, the genotype x year interaction was no significant but the genotype x location interaction was significant, and the year x location and genotype x year x location interactions were highly significant. High hereditability values were estimated in all cases indicating (h2 <0.94). A high number of QTL (Quantitative Trait Loci) related to the CIE b* and CPC were identified, mostly located on chromosomes 1BL, 2AS, 4AL (2), 5AS, 5BL, 6AL (3), 7AS, 7AL, 7BS, 7BL. These QTL were differentially expressed among to the environments but most QTL were stably anchored between the same molecular markers in all the environments. This is the first report on the existence of a QTL associated to the CIE b* and CPC located on chromosome 7AS. The presence of the QTL 7BS, its homeolog region, was recently reported. According to the combined mapping analyses, significant QTL (4AL, 6AL, 7BL) also have interaction effects. The QTL found in chromosome 6AL (6AL.2) explained most of the phenotypic variation for both traits over the average from the six environments. Irrespective of gene Psy-B1, these results suggest the presence of an extra gene on 7BL affecting the CPC. Molecular markers flanking the chromosome regions found in the combined analysis in chromosomes 6AL (Barc113-wmc553), 7AS (BQ170462_176-Barc174), 7BS (Barc23a-Barc72) and 7BL (PsyB-cfa2257), showed to be useful for selecting both traits, CPC and CIE b*. The number of environments required for QTL mapping is a critical factor because the same genotypes showed different gene/QTL expression patterns in different environments. The principal genomic regions related to CIE b* value and CPC are essentially the same.

Le tocophérol, le phytostérol, le pourcentage de protéines des graines, l'huile et les teneurs en acides gras ont été mesurés dans une population de lignées recombinantes (RILS) de tournesol, cultivées sous conditions de sécheresse, irrigation et semis tardif. Une analyse génétique de QTL a été réalisée à partir de ces mesures, en utilisant une carte génétique basée sur des marques SSR et avec des gènes candidats (1) impliqués dans la voie métabolique de tocophérol et phytostérol, (2) des gènes codant des antioxydants enzymatiques, (3) des gènes liés à la sécheresse et (4) des gènes homologues à SEC14 chez Arabidopsis. Trois gènes candidats importants (VTE4, VTE2 et HPPD), qui codent pour des enzymes impliquées dans la biosynthèse du tocophérol, ont été cartographiés sur les groupes de liaison LG8 et LG14. Quatre SNPs sont identifiés pour PAT2, le gène homologue chez Arabidopsis SEC14, entre les deux parents (PAC2 et RHA266) et un SNP, identifié par alignement de séquences est converti en marqueur CAPS pour permettre l'analyse génotypique des RIL. Les gènes homologues à SFH3, HPPD, CAT et CYP51G1 ont été cartographiés grâce à la mise au point de marqueurs dominants, tandis que des marqueurs co-dominants ont permis la cartographie des gènes homologues à SEC14-1, VTE4, DROU1, POD, SEC14-2 et AQUA. Les gènes POD, CAT et GST, codant pour des antioxydants enzymatiques, ont également été cartographiés sur les groupes de liaison 17, 8 et 1, respectivement. Le QTL majeur pour la teneur en tocophérol a été identifié sur le groupe de liaison 8, qui explique 59,5% de la variation phénotypique (6.TTC.8). Il colocalsie également avec le QTL identifié pour la teneur en phytostérol (7.TPC.8). Sous condition de semis tardif, un QTL spécifique de la teneur en acide palmitique a été identifié sur le groupe de liaison 6 (PAC-LS.6). Il est situé entre les marqueurs ORS1233 et SSL66_1. Les QTLs pour le pourcentage d'huile de graines et la teneur en acide stéarique colocalisent sur les groupes de liaison 10 (PSO-PI.10 et SAC-WI.10) et 15 (PSO-PI.15 et SAC-LS.15). Sept QTLs associés à teneur en acides palmitique, stéarique, oléique et linoléique sont identifiés sur le groupe de liaison 14. Ils sont liés à l’homologue du gène HPPD. Par ailleurs, les caractères agronomiques tels que les jours du semis à la floraison, la hauteur des plantes, le rendement et la morphologie foliaire ont été étudiés. Des analyses association génétique ont permis d’identifier des QTLs intérêts sur les groupes de liaison 2, 10 et 13 pour les caractères étudiés, d’autres QTLs identifies sur les groupes de liaison 9 et 12 mettent en avant l'importance de ces régions génomiques pour les caractères de morphologie foliaire. Nous avons finalement identifié des marqueurs AFLP et quelques gènes candidats liés aux caractères impliqués dans la qualité des graines sous conditions irriguée et stress hydrique chez une population de mutants (M8). Deux lignées mutantes, M8-826-2-1 et M8-39-2-1, produisent un niveau significativement élevé d'acide oléique peuvent être utilisées dans les programmes de sélection en raison de la haute stabilité à l'oxydation et des propriétés cardiovasculaire apportés par l’acide oléique qu’elles produisent. L'augmentation du niveau de tocophérol dans les lignées mutantes, M8-862-1N1 et M8-641-2-1, est justifiée par le polymorphisme observé pour le gène, MCT, impliqué dans la voie métabolique du tocophérol. Le marqueur le plus important pour le contenu en tocophérol total est E33M50_16 qui explique 33,9% de la variation phénotypique. Un des gènes candidats les plus importants concernant la biosynthèse des acides gras, FAD2 (FAD2-1), est lié à la teneur en acides oléique et linoléique. Il explique plus de 52% de la variation phénotypique. / The genetic control of tocopherol, phytosterol, percentage of seed protein, oil and fatty acids content in a population of recombinant inbred lines (RILs) of sunflower under various conditions are studied through QTL analysis using genetic-linkage map based on SSR markers and introducing some important tocopherol and phytosterol pathway-related genes, enzymatic antioxidant-related genes, droughtresponsive family genes and Arabidopsis SEC14 homologue genes. Three important candidate genes (HPPD, VTE2 and VTE4), which encode enzymes involved in tocopherol biosynthesis, are mapped to linkage group 8(LG8) and LG14. One of the most important candidate genes coding for sterol methyltransferase II (SMT2) enzyme is anchored to LG17 by CAPS marker. Four SNPs are identified for PAT2, Arabidopsis Sec14 homologue gene, between two parents (PAC2 and RHA266). PAT2 is assigned to LG2 by CAPS marker. Squalene epoxidase (SQE1) is also assigned to LG15 by InDel marker. Through other candidate genes, POD, CAT and GST encoding enzymatic antioxidants are assigned to LG17, LG8 and LG1, respectively. The major QTL for total tocopherol content on linkage group 8 accounted for 59.5% of the phenotypic variation (6.TTC.8), which is overlapped with the QTL of total phytosterol content (7.TPC.8). Under late-sowing condition, a specific QTL of palmitic acid content on linkage group 6 (PAC-LS.6) is located between ORS1233 and SSL66_1 markers. Common chromosomic regions are observed for percentage of seed oil and stearic acid content on linkage group 10 (PSO-PI.10 and SACWI. 10) and 15 (PSO-PI.15 and SAC-LS.15). Overlapping occurs for QTLs of oleic and linoleic acids content on linkage groups 10, 11 and 16. Seven QTLs associated with palmitic, stearic, oleic and linoleic acids content are identified on linkage group 14. These common QTLs are linked to HPPD homologue, HuCL04260C001. QTLs controlling various traits such as days from sowing to flowering, plant height, yield and leaf-related traits are also identified under well-, partial-irrigated and late-sowing conditions in a population of recombinant inbred lines (RILs). The results do emphasis the importance of the role of linkage group 2, 10 and 13 for studied traits. Genomic regions on the linkage group 9 and 12 are important for QTLs of leaf-related traits in sunflower. We finally identified AFLP markers and some candidate genes linked to seed-quality traits under well-irrigated and water-stressed conditions in gammainduced mutants of sunflower. Two mutant lines, M8-826-2-1 and M8-39-2-1, with significant increased level of oleic acid can be used in breeding programs because of their high oxidative stability and hearthealthy properties. The significant increased level of tocopherol in mutant lines, M8-862-1N1 and M8- 641-2-1, is justified by observed polymorphism for tocopherol pathway-related gene; MCT. The most important marker for total tocopherol content is E33M50_16 which explains 33.9% of phenotypic variance. One of the most important candidate genes involving fatty acid biosynthesis, FAD2 (FAD2-1), is linked to oleic and linoleic acids content and explained more than 52% of phenotypic variance.

La recherche de QTL par régression des phénotypes sur les probabilités de transmission (modèle Haley-Knott) est une méthode très largement utilisée quand on dispose de grandes familles phénotypées par des caractères gaussiens. L'objectif de cette thèse d'un point de vue méthodologique, est de proposer une méthode de détection de QTL qui prend en compte des effectifs de familles petits d'une part, et l'existence de caractères discrets d'autre part. Ainsi, nous proposons, pour répondre à la première question, une approche de détection de QTL intégrant dans le calcul du mérite génétique des individus marqués, les performances calculées sur n générations de descendants. L'obtention d'un mérite génétique dérégressé comme substitut de phénotypes, proposé notamment par Weller et al (1990) et Tribout et al (2008), est donc généralisée. Ensuite, sont présentés les résultats de comparaisons d'un modèle supposant la normalité des données à un modèle à seuils faisant l'hypothèse d'une distribution continue sous jacente à la distribution observée dans la détection de QTL des caractères discrets. Nous démontrons ici que le modèle discret est plus précis et plus puissant quand le caractère étudié possède trois modalités distribuées de façon déséquilibrée dans la population.Dans la deuxième partie de la thèse, l'analyse des données du protocole GENECAN a été réalisée. Il s'agit d'identifier les régions du génome ou locus à caractère quantitatif (QTL), associées à des caractères d'intérêt mesurés sur des canards mulards gavés. Le canard mulard est un hybride interspécifique obtenu par croisement d'une cane commune (Anas platyrhynchos) et d'un canard de Barbarie (Cairina moschata). Trois cents quarante deux canes communes conçues en back-cross (BC) ont été générées par croisement d'une lignée de canard Kaiya et d'une lignée de canard Pékin lourd. Ces femelles BC ont été accouplées avec des canards de Barbarie pour produire 1600 canards mulards sur lesquels sont effectuées des mesures de croissance, de métabolisme au cours de la période de croissance et du gavage, d'aptitude au gavage et de qualités du magret et du foie gras. La valeur phénotypique des femelles BC marquées a été estimée, pour chaque caractère, comme étant la valeur moyenne des phénotypes de sa progéniture et pondérée par un coefficient de détermination (CD) fonction du nombre de descendants et de l'héritabilité du caractère étudié. Une carte génétique de 91 marqueurs microsatellites réparties sur 16 groupes de liaison (GL) et couvrant un total de 778 cM a été utilisée. Dans le cadre de l'analyse uni-caractère, vingt-deux QTL significatifs à 1% au niveau du chromosome ont été cartographiés. Ces QTLs sont pour la plupart impliqués dans la variabilité de la qualité du magret et du foie gras. Les zones chromosomiques d'intérêt, identifiées dans le cadre de cette étude devront dans le futur, être densifiées en marqueurs pour faire l'objet d'une cartographie fine. / QTL detection using the regression of phenotypes on transmission probability is largely used when large families phenotyped for Gaussian trait are available. The aim of this thesis from a methodological point of view, is to propose a method for detection of QTL that takes into account the small number of families on the one hand, and the existence of discrete traits on the other. Thus, we propose to answer the first question, an QTL detection approach, integrating in the calculation of genetic merit of genotyped individuals, the performances calculated over n generations of descendants. The use of a ‘de-regressed proof' as a phenotype to be analysed, proposed by Weller et al. (1990) and Tribout et al. (2008) is generalized. Next, we present the results of comparisons of a model assuming normality of the data to a thresholds model assuming a continuous distribution underlying the observed distribution in the QTL detection of discrete traits. Here we demonstrate that the discrete model is more accurate and more powerful when the studied trait has three modalities distributed unevenly in the population.In the second part of the thesis, the data analysis of GENECAN protocol was performed. This is to identify genomic regions or quantitative trait locus (QTL) associated with interest traits measured on over-feed mule ducks. The mule duck is an hybrid duck from a female Common duck (Anas Platyrhynchos) and a Muscovy drake (Cairina moschata). Three hundred forty two common ducks designed by back-cross (BC) were generated by crossing a line of Kaiya duck and a heavy line of Pekin duck. These BC females were mated with Muscovy ducks to produce 1600 mules ducks which undergo measures of growth, metabolism during the growth and over-feeding periods, over-feeding, of breast muscle and fatty liver qualities. The phenotypic value of genotyped BC females was estimated for each trait as the average phenotypes of their offspring and weighted by a coefficient of determination (CD) function on the number of offspring and heritability of the studied trait. The genetic map comprised 91 microsatellite markers aggregated into 16 linkage groups (LG) and representing 778 cM. For the uni-trait analysis, twenty-two QTL significant at 1% threshold in chromosome-wide have been mapped. These QTLs are mostly involved in the variability of the breast muscle and fatty liver qualities. Chromosomal regions of interest identified in the framework of this study should be in the future be densified to markers to do the fine mapping.

L'objectif de cette thèse était d'analyser le déterminisme génétique de la croissance foliaire (LER) du maïs (Zea mays L.). Nous avons combiné plusieurs approches visant à (i) résumer l'information génétique tirée de trois populations de cartographie de QTL (une d'origine tropicale et deux tempérées), (ii) tester l'effet de l'introgresssion de diversité allélique dans les QTL les plus prometteurs, (iii) évaluer jusqu'à quel point les QTL de croissance foliaire affectent la croissance d'autres organes de la plante (iv) disséquer des QTLs d'intérêt par cartographie fine ou génétique d'association locale. La première partie de ce travail concerne le déterminisme génétique de la croissance foliaire maximale (LERmax) évaluée dans des conditions optimales la nuit. LERmax, telle que mesurée en plateforme de phénotypage, partage dans une forte proportion, son contrôle génétique avec la croissance d'autres organes. Des QTL qui affectaient LERmax ou/et la croissance d'autres organes ont alors été disséqués. Une région génomique a été cartographiée avec 23 lignées quasi-isogéniques (NILs) séquentiellement introgressées dans les bins 1.10-11, réduisant ainsi l'intervalle de confiance du QTL d'un facteur 3. Une seconde région génomique a été analysée par une méthode innovante fondée sur une étude d'association ciblée sur une région génomique dans un série allélique générée par introgression de 62 allèles donneurs tirés des lignées parentales d'hybrides cultivés et de populations historiques de maïs d'Amérique Latine dans une lignée élite. L'étude d'association dans cette région relativement petite révèle plus de polymorphismes causaux qu'attendus (six SNP en faible déséquilibre de liaison vs trois QTL consensus).La seconde partie de ce travail considère la sensibilité de la croissance foliaire à la demande évaporative et au déficit hydrique du sol. Un détermisme génétique commun aux deux sensibilités a été mis en évidence par méta-analyse de QTLs initialement détectés dans trois populations en ségrégation et par le test de NILs. Huit métaQTL situés dans quatre régions génomiques ont été testés avec 6 à 17 allèles introgressés pour identifier des NILs qui manifestaient les plus forts effets sur le phénotype. Nous avons initié une cartographie fine dans une de ces régions génomiques à partir d'une populations de recombinants issues de l'introgresssion d'un donneur d'origine tropicale dans B73. / The objective of this thesis was to analyze the genetic control of the Leaf Elongation Rate (LER) of maize (Zea mays L.). We combined approaches that (i) summarize the QTL information of three mapping populations (one tropical, two temperate), (ii) tested the impact of the introgression of allelic diversity at most promising QTLs, (iii) test to what extent QTLs of LER affect different traits (iv) dissect QTLs of interest by fine mapping or local association mapping.The first part of this document focuses on the genetic control of maximum LER (LERmax) measured in near-optimal conditions during the night. LERmax, as measured in a phenotyping platform, shares an appreciable proportion of its genetic control with the growth abilities of other organs. QTLs affecting LERmax and/or the growth of other organs were therefore dissected. One genomic region was fine-mapped with 23 Near-Isogenic Lines (NILs), sequentially introgressed in the bins 1.10-11, resulting in a reduction of the confidence interval by a factor 3. A second genomic region was analysed after the development of an innovative method of local association mapping on a collection of NILs, introgressed with 62 donor parents from historical populations from different altitude and latitudes in Latin America. This relatively small region harbors more causal polymorphisms than expected (six associated markers in low linkage disequilibrium vs three cQTLs).The second part focuses on the sensitivities of LER to evaporative demand or to soil water deficit. The two sensitivities share a large part of their genetic control as demonstrated by a metaQTL analysis on three mapping populations and the test of NILs. Eight metaQTLs in four genomic regions were tested with 6 to 17 different alleles to find the NILs that best impact the phenotype. We started a fine mapping on one genomic region by using one population of NILs involving a tropical donor.

Caracteres complexos são aqueles determinados por muitos genes e que apresentam variação contínua. Em uma população, a variação herdável dos caracteres complexos não é independente, e pares de caracteres podem ser mais ou menos correlacionados entre si. O nível e o padrão da associação entre caracteres determina como o fenótipo da população se comporta perante os processos evolutivos. A associação entre caracteres pode tanto facilitar a evolução em algumas direções do espaço fenotípico quanto restringir a evolução em outras, pois caracteres mais associados entre si tendem a evoluir de forma conjunta. O padrão de associação entre caracteres pode ser representado pela matriz de covariância genética aditiva, que descreve o padrão variacional resultante da interação do mapa genótipo-fenótipo e de todos os processos de desenvolvimento que levam desde a informação contida no material genético até o indivíduo. Tanto o mapa genótipo-fenótipo quanto o padrão de covariação genético também apresentam variação herdável, e portanto podem ser alterados pelos processos evolutivos e mudar entre gerações. Esse processo estabelece uma interação de mão dupla entre evolução e covariação, na qual a covariação afeta o resultado dos processos evolutivos e os processos evolutivos afetam a covariação. Nesta tese, nós exploramos como os efeitos genéticos interagem para formar o padrão de covariação, e como esses efeitos e covariação evoluem sob seleção natural. Para isso, nós trabalhamos com três populações experimentais de camundongos que foram sujeitas a regimes de seleção artificial e, utilizando diferentes tipos de caracteres, procuramos entender como a covariação se estabelece e como ela é afetada pela seleção. No primeiro experimento, estudamos o padrão de covariação de caracteres cranianos em linhagens selecionadas para aumento e diminuição do tamanho corporal, e observamos que a seleção para tamanho altera os caracteres do crânio e a covariação entre eles. A seleção direcional diminui a variação total do crânio, mas também aumenta a proporção de variação na direção de seleção, potencialmente facilitando uma nova resposta seletiva na mesma direção. Esse resultado implica que a variação presente em uma população pode ser moldada pela sua história evolutiva de forma adaptativa. No segundo experimento utilizamos uma população intercruzada, criada a partir linhagens selecionadas para aumento e diminuição do tamanho corporal, para identificar regiões genômicas envolvidas na determinação da curva de crescimento. Utilizando estimativas dos efeitos genotípicos nos fenótipos de crescimento, nós pudemos prever os fenótipos das linhagens ancestrais utilizando apenas informação da população intercruzada, e também construir estimativas de qual seria a covariação entre os caracteres de crescimento para cada tipo de efeito genético. Além disso, relacionamos a distribuição dos efeitos genéticos com a história evolutiva da população, mostrando que tanto a seleção quanto restrições internas do desenvolvimento interagem para determinar a distribuição de efeitos genéticos e, portanto, a covariação. No terceiro experimento, utilizamos seis linhagens de camundongos, que haviam sido selecionadas para alterações na curva de crescimento, para formar uma população intercruzada. Essa população apresentava uma enorme variação na sua curva de crescimento, e, utilizando técnicas de mapeamento genético, nós identificamos regiões genômicas envolvidas na determinação dessa variação fenotípica. Também desenvolvemos, para criar uma expectativa para a distribuição de efeitos genéticos nessa população, um modelo de simulação computacional da evolução dos efeitos genotípicos sob seleção. Os efeitos genéticos na população intercruzada apresentam um padrão mais complexo que o das simulações, e encontramos uma combinação de efeitos genéticos com padrões diferentes que interagem para gerar a covariação genética presente na população. Por fim, apresentamos uma revisão sobre a evolução da covariação genética e discutimos as consequências macroevolutivas das questões abordadas nos outros capítulos / Complex traits are defined as traits that are determined by many genes and that show continuous variation. In a population, the heritable variation of complex traits is not independent, and pairs of traits might be more or less correlated. The level and pattern of the association between traits determine how the phenotype of the population behaves when faced with evolutionary forces, like natural selection and genetic drift. The association between traits can both facilitate evolutionary change in some directions of the phenotype space and hinder change in other directions because tightly associated traits tend to evolve together. The pattern of association among traits can be represented by the additive genetic covariance matrix. This matrix describes the variational pattern that is the result of the interplay between the genotype-phenotype map and development, which together lead from the genetic information to the formation of the individual. Both the genotype-phenotype map and the genetic covariation also show heritable variation, and so are able to evolve and change between generations. This process establishes a feedback between evolution and covariation, in which covariation affects the outcome of the evolutionary process and is also shaped by evolution. In this thesis, we explore how genetic effects interact to create patterns of covariation, and how these effects and covariation change under natural selection. In order to do this, we use three experimental mice populations that were subjected to artificial selection regimes, and, using several types of complex traits, we study how covariation is established and how it evolves. In the first experiment, we use the covariation pattern of cranial traits measured in mice strains selected for the increase and decrease of body size. In these strains, we see that size selection altered the means of the cranial traits and the covariation between them. Directional selection reduces the total amount of genetic information, but in a non-uniform way. Some directions in phenotype space lose more variation than others, and, counter-intuitively, the direction of selection loses less variation. This leads to an increase in the proportion of variation that is in the direction of selection, potentially facilitating future evolutionary change in the same direction. This result shows that the covariation pattern in a population is shaped by its evolutionary history and can be adaptive. In the second experiment, we use an intercross population, created with two inbred mouse strains that were selected for increase and decrease in weight, to identify genomic regions involved in determining the growth curve of the individuals. Using estimates of the genetic effects on the growth traits, we were able to predict the phenotypes of the ancestral strains using only information from the intercross. We were also able to partition the genetic covariation into the contributions due to different types of genetic effects. We interpret the distribution of genetic effects in light of the evolutionary history of the population and show that the distribution of genetic effects, and of genetic covariation, is a consequence of the interaction between selection and development. In the third experiment, we create an intercross using six inbred mice strains that had been selected for different changes in their growth curve. This intercross shows large variation in growth curves, and, using genetic mapping techniques, we identify genomic regions involved in producing this phenotypic variation. To create an expectation for the distribution of genetic effects in this population, we develop a computer simulation model for the evolution of genetic effects under directional selection. The genetic effects in the population are more complex than in the simulation model, and we find that the genetic covariation between growth traits is created by the interaction among several different kinds of genetic effects. Finally, we present a review on the evolution of genetic covariation and discuss the macroevolutionary consequences of the themes we explore in the other chapters

Dezoito galinhas de uma população experimental utilizada em um cruzamento recíproco entre as linhagens de frangos de corte (TT) e de postura (CC) foram sequenciadas pela tecnologia de nova geração na plataforma Illumina com uma cobertura média de 10X. A descoberta de variantes genéticas foi realizada em uma região de locos de característica quantitativa (Quantitative Trait Locus, QTL), associado anteriormente com peso e percentagem de gordura abdominal no cromossomo 3 da galinha (GGA3), entre os marcadores microssatélites LEI0161 e ADL0371 (33,595,706-42,632,651 pb). O programa SAMtools foi utilizado na identificação de 136.054 SNPs únicos e 15.496 INDELs únicas nos 18 animais sequenciados e após a filtragem das mutações, 92.518 SNPs únicos e 9.298 INDELs únicas foram mantidas. Uma lista de 77 genes foi analisada buscando genes relacionados ao metabolismo de lipídios. Variantes localizadas na região codificante (386 SNPs e 15 INDELs) foram identificadas e associadas com vias metabólicas importantes. Variantes nos genes LOC771163, EGLN1, GNPAT, FAM120B, THBS2 e GGPS1 foram identificadas e podem ser responsáveis pela associação do QTL com a deposição de gordura na carcaça em galinhas. / Eighteen chickens from a parental generation used in a reciprocal cross with broiler and layer lines were sequenced by new generation technology with an average of 10-fold coverage. The DNA sequencing was performed by Illumina next generation platform. The genetic variants discovery was performed in a quantitative trait loci (QTL) region which was previously associated with abdominal fat weight and percentage in chicken chromosome 3 (GGA3) between the microsatellite markers LEI0161 and ADL0371 (33,595,706-42,632,651 bp). SAMtools software was used to detect 136,054 unique SNPs and 15,496 unique INDELs for the 18 chickens, and after quality filtration 92,518 unique SNPs and 9,298 unique INDELs were retained. One list of 77 genes was analised and genes related to lipid metabolism were searched. Variants located in coding region (386 SNPs and 15 INDELs) were identified and associated with important metabolic pathways. Loss of functional variants in the genes LOC771163, EGLN1, GNPAT, FAM120B, THBS2 and GGPS1 may be responsible for the QTL associated with fat deposition in chicken.

Muitas são as características quantitativas que são, significativamente, influenciadas por fatores genéticos, em geral, existem vários genes que colaboram para a variação de uma ou mais características quantitativas. As informações ausentes a respeito dos genótipos nos marcadores moleculares é um problema comum em estudo de mapeamento genético e, por conseguinte, no mapeamento dos locus que controlam estas características fenotípicas (QTL). Os dados que não foram observados ocorrem, principalmente, devido a erros de genotipagem e de marcadores não informativos. Para solucionar este problema foi utilizado o método do modelo oculto de Markov para inferir estes dados. Os métodos de acurácias evidenciaram o sucesso da aplicação desta técnica de imputa- ção. Uma vez imputado, na inferência bayesiana estes dados não serão mais tratados como uma variável aleatória resultando assim, numa redução no espaço paramétrico do modelo. Outra grande dificuldade no mapeamento de QTL se deve ao fato de que não se conhece ao certo a quantidade destes que influenciam uma dada característica, fazendo com que surjam diversos problemas, um deles é a dimensão do espaço paramétrico e, consequentemente, a obtenção da amostra a posteriori. Assim, com o objetivo de contornar este problema foi proposta a utilização do método Monte Carlo via cadeia de Markov com Saltos Reversíveis, uma vez que este permite flutuar, entre cada iteração, modelos com diferentes quantidades de parâmetros. A utilização da abordagem bayesiana permitiu detectar cinco QTL para a característica estudada. Todas as análises foram implementadas no programa estatístico R. / There are many quantitative characteristics which are significantly influenced by genetic factors, in general, there are several genes that contribute to the variation of one or more quantitative trait. The missing information about the genotypes in molecular markers is a common problem in studying genetic mapping and therefore the mapping of loci that control these phenotypic traits (QTL). The data were not observed occur mainly due to errors in genotyping and uninformative markers. To solve this problem the method of occult Markov model to infer this information was used. Techniques accuracies demonstrated the successful application of this technique of imputation. Once allocated, in the Bayesian inference this data will no longer be treated as a random variable thus resulting in a reduction in the parameter space of the model. Another great difficulty in mapping QTL is due to the fact that no one knows exactly the amount of these which influence a given characteristic, so that several problems arise, one of them is dimension of the parameter space and, consequently, obtaining the sample a posterior. Thus, in order to solve this problem using the method via Monte Carlo Markov chain Reversible Jump was proposed, since this allows fluctuate between each iteration, models with different numbers of parameters. The use of the Bayesian approach allowed five QTL detected for the studied trait. All analyzes were implemented in the statistical software R.