No description available.

L’agroécologie a des attentes fortes en termes de recherche sur les concepts et sur les méthodes de sélection et de gestion de la diversité. En effet, la prise en compte de la diversité des environnements et des pratiques spécifiques à chaque système agroécologique appelle des approches plus décentralisées, c’est à dire une meilleure prise en compte des interactions génotype × environnement, et associant les savoirs empiriques des praticiens aux connaissances scientifiques.L’objectif de cette thèse est de développer une méthodologie de la sélection participative pour le blé tendre, basée sur la décentralisation et la co-construction entre paysans, associations du Réseau Semences Paysannes et chercheurs de l’équipe DEAP de l’INRA du Moulon. Cette approche vise à créer des variétés-populations adaptées aux environnements et aux pratiques des paysans, à développer des méthodes et des outils opérationnels pour la gestion et la sélection de la biodiversité cultivée à la ferme et à renforcer l’apprentissage et l’autonomie des paysans en matière de sélection.Ce travail a contribué à atteindre ces objectifs, notamment les deux derniers, en se basant sur une approche interdisciplinaire dans les champs de la statistique, génétique quantitative, génétique des populations, bio-informatique et sociologie.Les objectifs ont été en partie atteints avec la création de méthodes et d’outils : fiches de suivi de la culture, base de données permettant de gérer les relations entre lots de semences dans un large réseau d’acteurs, dispositifs expérimentaux à la ferme et méthodes statistiques permettant de prendre en compte le déséquilibre des essais à la ferme, en profitant du large réseau d’expérimentation, afin de réaliser des comparaisons de moyennes dans les fermes et d’analyser les interactions génotype × environnement dans le réseau de fermes, programme informatique qui permet de créer un dossier avec des résultats personnalisés pour chaque paysan, livret technique sur la sélection participative.Une analyse moléculaire et phénotypique montre que le programme de sélection participative s’insère dans la gestion in-situ des ressources génétiques par la création de nouvelles variétés-populations et par leur gestion dans un large réseau de fermes regroupant une grande diversité d’environnements et de pratiques.Les paysans ont renforcé leur apprentissage et leur autonomie. Ils étaient acteurs du programme et leurs pratiques ont évolué avec, par exemple, un nombre croissant de populations évaluées sur le réseau, de sélections et d’échanges de semences entre paysans au fur et à mesure que le projet avançait.Deux innovations émergent de ce projet : organisationnelle et génétique. L’innovation organisationnelle correspond au mode de fonctionnement du projet, basé sur la décentralisation de la sélection directement dans les fermes et la co-construction entre les acteurs et leur mise en réseau. Les innovations génétiques découlent de ce mode d’organisation : une large diversité génétique et phénotypique a été créée, est maintenue et évaluée dans les fermes et de nouvelles variétés populations, issues du programme, sont utilisées par des paysans dans leurs fermes.La méthodologie que nous avons développée est évolutive, flexible et adaptable. Elle place les paysans au coeur de la sélection et de la gestion des ressources génétiques. Ce projet participe à la transition vers un nouveau système semencier adapté à l’agroécologie / Taking into account the diversity of environment and practices to each agroecological system needs decentralized approaches, which means taking into account genotype environment interactions, and associate empirical knowledge from practitioners to scientific knowledge from researchers.The objectives of this PhD is to develop a participatory plant breeding methodology on bread wheat, based on decentralization and co-construction between farmers, NGOs and reasearch team DEAP from INRA Le Moulon.This approach aims to create population-varieties adapted to environments and farmers' practices, to develop methods and tools for breeding and cultivated biodiversity management and to reinforce learning and autonomy of farmers for breeding.This work contributed to reach these objectives, especially the last two, based on an interdisciplinary approach in statistics, quantitative genetic, populations genetic, bio-informatic and sociology.Objectives were partly reach with creation of methods and tools : monitoring sheet of the crop, data base which manage relation between seed lots in a large network of actors, experimental design on farm and statistical methods that take into account the disequilibrium of trials on farm, taking advantage of the large network, in order to make mean comparisons on farm and analyse genotype x environments interactions in the network of farms, software which create a folder with personalised results for each farmer, technical booklet on participatory plan t breeding.A phenotypic and molecular analysis shows that the programme takes part in in-situ management of genetic resources by creating new population-varieties and their management in a broad network of farms.Moreover, farmers reinforce their learning and their autonomy. They were actors of the programme and their practices evolved with, for example, an increasing number of populations evaluated on the network, of selection, of seed exchange between farmers as the project go on.Two innovations emerged from this project : organisational and genetic. The organisational innovation correspond to the way the project worked, based on decentralisation of the selection directly on farms and co-construction between actors and their networked. Genetic innovations are the consequence of this organisation : a brad genetic and phenotypic diversity has been created, and maintain and evaluated on farms and new population-varieties, coming from the project, are used by farmers on their farms.The methodology that we developed is evolutive, flexible and adaptable. It put farmers in the heart of selection and biodiversity management. This project contribute to a transition toward a new seed system adapted to agroecology.

The use of molecular markers allows an increase in efficiency of the selection as well as better understanding of genetic resources in breeding programs. However, with the increase in the number of markers, it is necessary to process it before it can be ready to use. Also, to explore Genotype x Environment (GE) in the context of genomic prediction some covariance matrices needs to be set up before the prediction step. Thus, aiming to facilitate the introduction of genomic practices in the breeding program pipelines, we developed two R-packages. The former is called snpReady, which is set to prepare data sets to perform genomic studies. This package offers three functions to reach this objective, from organizing and apply the quality control, build the genomic relationship matrix and a summary of a population genetics. Furthermore, we present a new imputation method for missing markers. The latter is the BGGE package that was built to generate kernels for some GE genomic models and perform predictions. It consists of two functions (getK and BGGE). The former is helpful to create kernels for the GE genomic models, and the latter performs genomic predictions with some features for GE kernels that decreases the computational time. The features covered in the two packages presents a fast and straightforward option to help the introduction and usage of genome analysis in the breeding program pipeline. / O uso de marcadores moleculares permite um aumento na eficiência da seleção, bem como uma melhor compreensão dos recursos genéticos em programas de melhoramento. No entanto, com o aumento do número de marcadores, é necessário o processamento deste antes de deixa-lo disponível para uso. Além disso, para explorar a interação genótipo x ambiente (GA) no contexto da predição genômica, algumas matrizes de covariância precisam ser obtidas antes da etapa de predição. Assim, com o objetivo de facilitar a introdução de práticas genômicas nos programa de melhoramento, dois pacotes em R foram desenvolvidos. O primeiro, snpReady, foi criado para preparar conjuntos de dados para realizar estudos genômicos. Este pacote oferece três funções para atingir esse objetivo, organizando e aplicando o controle de qualidade, construindo a matriz de parentesco genômico e com estimativas de parâmetros genéticos populacionais. Além disso, apresentamos um novo método de imputação para marcas perdidas. O segundo pacote é o BGGE, criado para gerar kernels para alguns modelos genômicos de interação GA e realizar predições genômicas. Consiste em duas funções (getK e BGGE). A primeira é utilizada para criar kernels para os modelos GA, e a última realiza predições genômicas, com alguns recursos especifico para os kernels GA que diminuem o tempo computacional. Os recursos abordados nos dois pacotes apresentam uma opção rápida e direta para ajudar a introdução e uso de análises genômicas nas diversas etapas do programa de melhoramento.

Submitted by Verena Pereira (verenagoncalves@uefs.br) on 2018-07-17T21:59:57Z No. of bitstreams: 1 Tese Final_ Danillo Olegario.pdf: 8604713 bytes, checksum: 2e0a9905d4de264fbbd3a8997c3babb3 (MD5) / Made available in DSpace on 2018-07-17T21:59:57Z (GMT). No. of bitstreams: 1 Tese Final_ Danillo Olegario.pdf: 8604713 bytes, checksum: 2e0a9905d4de264fbbd3a8997c3babb3 (MD5) Previous issue date: 2018-03-16 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES / The goals of this work were to compare and evaluate statistical methods of analysis of adaptability and stability for total protein, minerals and grain yield in cowpea lines, as well as retention of these nutrients after cooking to enable recommendation and registration of new cultivars for the Brazilian semiarid. Forty-four genotypes, distributed in two experiments of different population densities, were evaluated in seven environments, in the states of Bahia, Cear?, Pernambuco and Piau?. Initially, a comparison was made of ten statistical methods of adaptability and stability using only the production data. Spearman correlation coefficients were estimated between the methods. For the assessments of adaptability and stability of the genotypes the methods of Eberhart and Russell, Lin and Binns and the Model of main additive effects and multiplicative interaction - AMMI were applied. To evaluate the effect of cooking, 24 genotypes were used, including ten lines, five commercial cultivars and nine landraces maintained by farmers. The seed protein content was quantified through the Kjeldahl method. and the minerals by the AOAC standard procedure. The statistical analyzes for the experimental designs were performed in the SAS by the GLM procedure. The AMMI method is most suitable due to the graphical arrangement of both the genotypes and the environments and high correlation with others evaluated methods. The use of highly correlated methods is not recommended, as in the cases of Plaisted and Peterson and Wricke and the methods of Annicchiarico and Lin and Binnsthat show strong association and produce similar genotypic classifications. The use of one of them incombination with that of Eberhart and Russell or AMMI can add information to the stability analysis. Statistically significant differences in the genotype as well as in the genotype?environment interaction for minerals, proteins and grain yield were observed in all the assays. The lines presenting the highest protein contents showed the lowest grain yields, and it indicated the prominent ?phenotypic cost? of protein in overall cowpea seed production. The methods by Eberhart and Russell, Lin and Binns and AMMI showed similar results for selection of superior genotypes. For the protein content, the lines C3Q, C3M, C2S and C1J presented grain yields equal to or greater than the general mean of the experiments and average of 27% of protein. For iron and zinc contents, the line C4I and T16_2R presented grain yield equal to or greater than the general mean of the experiments, with mean values of Fe and Zn 15% higher than the values of the evaluated cultivars. For the potassium and calcium contents, the C4I and C3O strains showed grain yield equal to or greater than the general average, with K and Ca means higher than the values of the cultivars evaluated. All the lines mentioned above presented broad adaptability and good stability in the series of evaluated environments showing great potential to be released as new cultivars for the Brazilian semiarid. Cooking had a reduced effect on protein, potassium, calcium, iron and zinc contents in the cowpea beans. The CPCR3F6L17 obtained high grain yield and high levels of protein, potassium, iron and zinc, after cooking, being promising for the region studied / Os objetivos deste trabalho foram comparar e avaliar m?todos estat?sticos de an?lise de adaptabilidade e estabilidade para o teor de prote?nas totais, minerais e produ??o de gr?os em linhagens de feij?o-caupi, bem como a reten??o desses nutrientes ap?s a coc??o, para possibilitar a recomenda??o e registro de novos cultivares para o semi?rido brasileiro. Quarenta e quatro gen?tipos, distribu?dos em dois experimentos de diferentes densidades populacionais, foram avaliados em sete ambientes, nos estados da Bahia, Cear?, Pernambuco e Piau?. Inicialmente, foi realizado acompara??o de dez m?todos estat?sticos de adaptabilidade e estabilidade utilizando apenas dados de produ??o. Coeficientes de correla??o Spearman foram estimados entre os m?todos. Para as avalia??es de adaptabilidade e estabilidade dos gen?tipos aplicou-se os m?todos de Eberhart e Russell, Lin e Binns e o Modelo de efeitos aditivos principais e intera??o multiplicativa - AMMI. Para avaliar o efeito da coc??o, foram utilizados 24 gen?tipos, sendo dez linhagens, cinco cultivares comerciais e nove variedades mantidas por agricultores.A quantifica??o de prote?na total foi realizada pelo m?todo de Kjeldhal, e os minerais pelo procedimento padr?o da AOAC. As an?lises estat?sticas para os delineamentos experimentais foram efetuadas no SAS pelo procedimento GLM. O m?todo AMMI ? o mais indicado devido a disposi??o gr?fica e alta correla??o com os demais m?todos. A utiliza??o de m?todos que apresentam elevada correla??o, como os m?todos Plaisted e Peterson e Wricke, Annicchiarico e Lin e Binns, que mostram forte associa??o, e produzem classifica??es genot?picas similares, n?o ? recomendado. O uso de um deles em combina??o com o de Eberhart e Russell ou AMMI pode agregar informa??o ? an?lise de estabilidade. Foram observadas diferen?as estat?sticas significativas dos quadrados m?dios de tratamentos, dos ambientes e das intera??es ambientes x tratamentos para minerais, prote?nas e produ??o de gr?os. As linhagens que apresentaram os maiores teores de prote?na e minerais apresentaram produ??es de gr?os abaixo da m?dia geral, nos dois experimentos. Os m?todos de Eberhart e Russell, Lin e Binns e o AMMI apresentam resultados semelhantes quanto a sele??o de gen?tipos superiores. Para o teor de prote?na, As linhagens C3Q, C3M, C2S e C1J apresentaram produ??es de gr?os igual ou superior a m?dia geral dos experimentos e m?dia de 27% de prote?na. Para os teores de ferro e zinco, a linhagem C4I e T16_2R apresentou produ??o de gr?os igual ou superior a m?dia geral dos experimentos, com valores m?dios de Fe e Zn, 15% superiores aos valores dos cultivares avaliados. Para os teores de pot?ssio e c?lcio, as linhagens C4I e C3O mostraram rendimento de gr?os igual ou maior do que a m?dia geral, com m?dia de K e Ca mais elevadas do que os valores dos cultivares avaliados. Todas as linhagens citadas anteriormente apresentaram ampla adaptabilidade e boa estabilidade na s?rie de ambientes avaliados mostrando grande potencial para serem lan?ados como novos cultivares para o semi?rido brasileiro. A coc??o teve efeito reduzido nos teores de prote?na, pot?ssio, c?lcio, ferro e zinco, nos gr?os de feij?o-caupi. A linhagem CPCR3F6L17 obteve alta produtividade de gr?os e elevados teores de prote?na, pot?ssio, ferro e zinco, ap?s a coc??o, mostrando-se promissora para a regi?o estudada

A produtividade de grãos em milho é um caráter altamente complexo e muito dependente das condições ambientais. Neste trabalho de pesquisa, buscouse identificar regiões cromossômicas relacionadas à produtividade de grãos em milho por meio de análises de regressão múltipla stepwise, em vários ambientes. Os dados genotípicos advém da genotipagem de linhagens S2 por 163 locos RFLP, enquanto os dados fenotípicos foram obtidos de experimentos com repetições instalados em 11 locais distintos, nos quais foram avaliados os topcrosses das linhagens com quatro testadores diferentes. Foram selecionadas marcas associadas ao caráter nos diferentes ambientes, posteriormente comparadas a fim de verificar a consistência na expressão de genes das regiões detectadas. De maneira geral, observou-se que a maioria delas é ambiente-específica. Após a seleção das marcas, foi aplicado um índice de seleção de linhagens, baseado nos valores bj da regressão múltipla. O índice é composto pelo valor próprio das linhagens em topcrosses e por uma medida da complementaridade genotípica entre linhagens. As melhores linhagens, segundo o índice de seleção, foram agrupadas por testador e verificouse que a coincidência de linhagens entre locais variou de 48,5% a 80,0%. O emprego de marcadores permitiu verificar que a interação genótipo x ambiente foi devida à alta inconsistência na manifestação das regiões cromossômicas responsáveis pela produtividade de grãos em milho ao longo dos locais. Porém, a coincidência entre as linhagens indica que, mesmo na presença de interação, foi possível selecionar linhagens generalistas. / Grain yield in maize is a complex trait, highly dependent on environmental conditions. Identification of consistent chromosomal regions across environments related to this trait is desirable in the context of marker assisted selection. This work was aimed at identifying regions associated with maize grain yield by stepwise multiple regression analysis in several environments. Maize S2 lines were genotyped by 163 RFLP loci and topcrossed to four different testers. These top crosses were evaluated in 11 locations. Grain yield associated markers were identified for each environment and the consistency of expression of associated genes was verified. Most of the chromosome regions were environment specific. After marker identification for each environment, a selection index was applied, which was composed by the performance of a line in topcross and a measure of genetic complementarity. The best lines were grouped by tester across locations and the coincidence of superior lines varied from 48.5% to 80.0%. The use of molecular markers showed that genotype x environment interaction was due to a high expression inconsistency of chromosome regions related to grain yield in maize across environment. Nonetheless, the coincidence between superior lines indicated the possibility to selecting lines with good performance along environments, even in the presence of genotype x environment interaction.

Submitted by (ana.araujo@ufrpe.br) on 2017-02-21T14:42:34Z No. of bitstreams: 1 Pedro Henrique Neves de Souza.pdf: 639543 bytes, checksum: c4f4efdf9da82b1b9d2aa60d34c8085f (MD5) / Made available in DSpace on 2017-02-21T14:42:34Z (GMT). No. of bitstreams: 1 Pedro Henrique Neves de Souza.pdf: 639543 bytes, checksum: c4f4efdf9da82b1b9d2aa60d34c8085f (MD5) Previous issue date: 2011-09-19 / Brazil is of the most traditional producers of cane sugar, whose culture is deployed in various soil types and under the influence of different climates, resulting in various types of environments with which it interacts. As for the agronomy, the culture of sugar cane has an excellent development, a result of the timing between the technology involved in growing and favorable climatic conditions. Through the methodology, this work aimed to study the micro-center of the state of Pernambuco, the behavior of clones and varieties of cane sugar in the breeding program of the Federal Rural University of Pernambuco - UFRPE / RIDESA analyzing aspects agro-industrial productivity, as well as assessing genotype x environment interactions, estimating the magnitude of some genetic parameters and based on the analysis of economic return, prepare a recommendation for use of superior materials. The experiment was conducted at the microregion area of Pernambuco Center, Plant Petribú premises, located in the municipality of Lagoa de Itaenga - PE during the agricultural years 2005/2006, 2006/2007 and 2007/2008. We used randomized complete block design with four replications, using 11 varieties and 15 clones have already been exploited commercially. Agrotechnological variables were evaluated tons of cane per hectare (TCH), tons of sugar per hectare (TPH), pol% corrected (PCC), fiber (FIB), purity (PZA), soluble solids (Brix) and total sugar recoverable (ATR). Analysis were performed by cutting individual and joint of the three cuts (plant cane, first ratoon crop and second ratoon crop) to the beginning, middle and end of harvest, also some estimating genetic parameters and their averages grouped by Scott & Knott test (1974). Finally were performed to analyze the economic return of the materials under study. / O Brasil é um dos mais tradicionais produtores de cana-de-açúcar, cuja cultura é implantada em vários tipos de solos e sob influência de diferentes climas, o que resulta em diversos tipos de ambientes com as quais ela interage. Quanto ao aspecto agronômico, a cultura da cana-de-açúcar apresenta um desenvolvimento excelente, resultado do sincronismo entre a tecnologia aplicada no cultivo e as condições climáticas favoráveis. Através da metodologia aplicada, este trabalho teve como objetivo estudar na microrregião Centro do Estado de Pernambuco, o comportamento de clones e variedades de cana-de-açúcar no programa de melhoramento genético da Universidade Federal Rural de Pernambuco – UFRPE / RIDESA, analisando aspectos relacionados à produtividade agroindustrial, bem como as interações genótipo x ambiente, estimando a magnitude de alguns parâmetros genéticos e com base na análise de retorno econômico, eleger materiais superiores. O experimento foi conduzido na zona canavieira da microrregião Centro de Pernambuco, na área agrícola da Usina Petribú, localizada no município de Lagoa de Itaenga - PE, durante os anos agrícolas 2005/2006, 2006/2007 e 2007/2008. Foi utilizado o delineamento em blocos casualizados completos com quatro repetições, utilizando 11 clones e 15 variedades comercialmente já exploradas. Foram avaliadas as variáveis toneladas de cana por hectare (TCH), toneladas de açúcar por hectare (TPH), pol % corrigido (PCC), fibra (FIB), pureza (PZA), teor de sólidos solúveis (BRIX) e açúcar total recuperável (ATR). Foram realizadas análises individuais por corte e conjunta dos três cortes (cana planta, soca e ressoca) para o início, meio e final de colheita, estimando também alguns parâmetros genéticos e suas médias agrupadas através do teste Scott & Knott (1974). Por fim foram realizadas análises de retorno econômico dos materiais em estudo.

This thesis is an investigation of the genetic basis of tocopherol accumulation in soybean (Glycine max [L.] Merrill) seeds. Soybean is the world’s most widely grown protein and oilseed crop and the principle source of vitamin E (tocopherols) as a supplement. Tocopherols (α-, β-, γ- and δ-isomers) are powerful antioxidants that contain human health benefits, including a decrease in the risk of lung cancer, heart disease and osteoporosis. The purpose of this research was to identify genetic and biochemical components affecting tocopherol accumulation in soybean seeds. The objectives were to: 1) investigate location and year effects on soybean seed tocopherol levels in the field; 2) determine environmental factors affecting soybean seed tocopherol levels under controlled conditions; 3) identify simple sequence repeat (SSR) markers that tag quantitative trail loci (QTL) for individual and total tocopherols; and 4) evaluate the potential role of VTE1, VTE3 and VTE4 genes in tocopherol accumulation using the candidate gene approach. Seventy nine recombinant inbred lines (RILs) derived from the cross between OAC Bayfield and OAC Shire were grown in the field at Elora, Woodstock and St. Pauls, ON, in 2009 and 2010. The tocopherol components were quantified using high performance liquid chromatography (HPLC). The results showed a significant (p < 0.001) genotype, environment and genotype x environment effect for each tocopherol component. It was discovered that a 2 x phosphate fertilizer (K2SO4 at 1.0M/150mL) and 30 ˚C temperature treatment increased each tocopherol component, whereas drought had no effects. Single marker analysis identified 42 QTL and interval mapping identified 26 QTL across 17 chromosomes. Significant two-locus epistatic interactions were found with a total of 122 and 152 in the 2009 and 2010 field seasons, respectively. The multiple locus models explained 18.4% to 72.2% with an average of 45.7% of the total phenotypic variation. The candidate gene approach using nucleotide sequences from the coding regions identified two single nucleotide polymorphisms (SNPs) in VTE1, five SNPs in VTE4 and none in VTE3. The SNPs were predicted to cause functional protein changes and the genes co-localized with some of the identified QTL. The results of this study provide a better understanding of the environmental factors and genetic mechanisms that influence the accumulation of tocopherols in soybean seeds. / Grain Farmers of Ontario, Vitamin Research Award

Au Sahel, la répartition des pluies irrégulière dans le temps et dans l’espace engendre une forte interaction variété x année et variété x lieu. Pour déterminer les variétés les plus productives en espérance en fonction des lieux, il faudrait de nombreuses années d’expérimentation en chaque lieu, ce qui prendrait beaucoup de temps.Une alternative est de maximiser la production prédite à l’aide d’un modèle de culture décrivant la croissance et le développement de cultures en interaction avec leurs conditions agro-environnementales.La production à maximiser est moyenne sur une distribution de probabilitéd’entrées environnementales, spécifique du lieu, alors que les paramètres variétaux qui maximisent cette production définissent un but de sélection que l’on appelle idéotype.Dans ce travail, nous voulons déterminer une carte d’idéotypes de sorgho. Nous sommes donc confrontés à un problème d’optimisation d’un modèle complexe. Une méthode classiquement utilisée dans ce contexte est la méthode Efficient Global Optimization (EGO), fondée sur un métamodèle de krigeage. Ici, une telle approche n'est pas adaptée. En effet, la distribution des entrées météorologiques suit un modèle stochastiquedont les paramètres varient continûment dans l’espace en suivant un gradient Nord-Sud. L’optimisation des paramètres variétaux est alors conditionnelle à ces paramètres de climat. D’autre part, la fonction à maximiser n’est connue que par un nombre limité de simulations, donc à une erreur près.Notre cadre de travail concerne donc l'optimisation conditionnelle d'une fonction bruitée. Les extensions existantes de l'algorithe $EGO$ ne prennent pas en charge ce cadre. Dans cette thèse, un nouveau critère pour l'optimisation conditionnelle d'une fonction bruitée est proposée et étudié. Une métaphore de l'optimisation conditionnelle est la recherche d'une ligne de crête. A partir de simulations sur des fonctions test, une étude des performances de ce nouveau critère est proposée, de même qu'une comparaison avec le critère habituellement utilisé pour la recherche de ligne de crête. Les résultats de cette étude montrent l'intérêt de notre critère.L'application à la cartographie d'idéotypes de sorgho a été testée sur l'espace couvert par le Sénégal, le sud du Mali et le Burkina Faso. Elle a consisté à maximiser le rendement espéré en fonction de 4 paramètres du modèle Samara : la longueur de la phase végétative, la longueur maximale des racines, le potentiel de réserve des tiges, et la mortalité des feuilles. Les résultats de cette optimisation recoupent en partie l'analyse de sensibilité menée sur ces mêmes paramètres. / In the Sahel region, the irregular rainfall distribution in time and space generates variety x year and variety x location interactions. Therefore, determining variety with the best expected yield would take many years of experimentation in each location.Alternatively, the best variety could be identified by maximizing the predicted yield using a crop simulation model that describes growth and development of a crop in interaction with agro-environmental conditions.The average yield depends on the probability distribution of environmental inputs, which is location specific, while the cultivar parameters that maximizethis yield define the ideotype, i.e. the selection target.In this work, we want to draw the map of sorghum ideotypes in Sub Saharan Africa. To face the problem of optimizing a complex model, an algorithm conventionally used in this context is the Efficient Global Optimization method (EGO), based on kriging as a surrogate model. Here, the distribution of meteorological inputs follows a stochastic model whose parameters varycontinuously in space along a North-South gradient. Consequently, the optimization of varietal parameters is conditional on these climate parameters. Moreover, the function to maximize is noisy, because expectation and quantilesare merely estimated with a limited number of simulations. We aimed at adapting the EGO algorithm to the conditional optimization of a noisy function. Extensions exist either for the optimization of noisy functions or for the conditional optimization of deterministic functions, ie the search for the values of a subset of parameters that optimize the function conditionally to the values taken by another subset, which are fixed. A metaphor for conditional optimization is the search for a crest line. No method has yet been developed for the conditional optimization of noisy functions: this is what we propose in this thesis. Testing this new method on test functions shows that, in case of a high level of noise on the function, the PEQI criterion that we propose is better than the PEI criterion usually implemented in such a situation.The application of this new optimization method sorghum ideotypes parameters mapping has been tested in the area covered by Senegal, southern Mali and Burkina Faso. It consisted in maximizing the expected yield with respect to 4 parameters of Samara model: vegetative phase length, maximum root length, stem reserve potential, and leaf mortality. The results of this optimization partly coincide with the sensitivity analysis conducted on these same parameters.

Linha de pesquisa: Melhoramento animal. / Submitted by Jos? Henrique Henrique (jose.neves@ufvjm.edu.br) on 2017-09-29T00:35:26Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) diego_coimbra_alcantara.pdf: 1887370 bytes, checksum: 3b18dc6adcb9630901b41a57c4a4a351 (MD5) / Approved for entry into archive by Rodrigo Martins Cruz (rodrigo.cruz@ufvjm.edu.br) on 2017-10-09T14:23:40Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) diego_coimbra_alcantara.pdf: 1887370 bytes, checksum: 3b18dc6adcb9630901b41a57c4a4a351 (MD5) / Made available in DSpace on 2017-10-09T14:23:40Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) diego_coimbra_alcantara.pdf: 1887370 bytes, checksum: 3b18dc6adcb9630901b41a57c4a4a351 (MD5) Previous issue date: 2016 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico (CNPq) / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior (CAPES) / Funda??o de Amparo ? Pesquisa do estado de Minas Gerais (FAPEMIG) / Objetivou-se com este estudo, caracterizar a intera??o gen?tipo x ambiente em codornas de corte, alimentadas com dietas contendo diferentes rela??es de triptofano:lisina, via normas de rea??o. Para realizar o experimento foram utilizadas duas linhagens de codornas de corte, LF1 e LF2. A ra??o fornecida no per?odo de crescimento, nascimento aos 21 dias de idade, continha uma rela??o de 0,20% de triptofano:lisina, e no per?odo final do crescimento, 22 a 35 dias, as codornas foram separadas em 5 tratamentos com as rela??es: 0,15; 0,20; 0,25; 0,30; 0,35% de triptofano:lisina. As aves foram pesadas aos 28 e 35 dias e abatidas aos 35 dias de idade. As caracter?sticas avaliadas foram os pesos pr?-abate, peso 28, peso 35, peso e rendimentos da carca?a, peito, coxa+sobrecoxa e as caracter?sticas de qualidade da carne. As an?lises foram realizadas utilizando modelos de regress?o aleat?ria considerando efeito fixo de sexo e o efeito aleat?rio gen?tico aditivo direto como fun??es das rela??es triptofano:lisina da dieta, admitindo-se homogeneidade de vari?ncia residual. Os resultados para desempenho, caracter?sticas de carca?a e caracter?sticas de qualidade da carne indicam que o coeficiente de regress?o do intercepto (b0), foi maior que o coeficiente de regress?o aleat?rio (b1). A herdabilidade e a vari?ncia gen?tica aditiva foram influenciadas pelas diferentes rela??es triptofano:lisina da dieta. Houve intera??o gen?tipo x ambiente para as caracter?sticas rendimento de carca?a, rendimento de peito e rendimento de coxa+sobrecoxa. As caracter?sticas de qualidade da carne CRA, b*, c* e H* (LF1) e CRA, PPC, L* e H* (LF2) apresentaram intera??o gen?tipo x ambiente. A intera??o gen?tipo x ambiente observada para as caracteristicas de rendimento e qualidade da carne indicam que a sele??o seja feita na rela??o triptofano:lisina em que as codornas ser?o criadas. / Disserta??o (Mestrado) ? Programa de P?s-Gradua??o em Zootecnia, Universidade Federal dos Vales do Jequitinhonha e Mucuri, 2016. / We animed of this experiment was to characterize the genotype x environment interaction in the production of meat quails, fed diets containing different tryptophan:lysine ratios, via reaction standards. Two lines of cutting quails, LF1 and LF2, were used to perform the experiment. The feed provided in the growing period, from birth to 21 days, contained a 0.20% tryptophan:lysine ratio and in the final growth period, from 22 to 35 days of age, quails were separated into 5 treatments containing the Ratios of 0.15; 0.20; 0.25; 0.30; 0.35% tryptophan: lysine. The birds were weighed at 28 and 35 days of age and slaughtered at 35 days of age. The evaluated characteristics were the pre-slaughter weights, weight 28, weight 35, weight and yields of the carcass, breast, thigh + sobrecoxa and the quality traitsof the meat. The analyzes were performed using random regression models considering fixed sex effect and direct additive genetic random effect as functions of the tryptophan:lysine ratios of the diet, assuming homogeneity of residual variance. The results for performance, carcass traitsand meat quality traitsindicate that the regression coefficient of the intercept (b0) was higher than the random regression coefficient (b1). Heritability and additive genetic variance were influenced by the different tryptophan:lysine ratios of the diet. There was no genotype x environment interaction for the traitsof carcass yield, yield of breast and yield of thigh + overcoat. The quality traits of the meat CRA, b *, c * and H * (LF1) and CRA, PPC, L * and H * (LF2) presented genotype x environment interaction. It is possible to conclude that for yield traitsthe selection can be made in the lowest tryptophan: diet lysine ratios. For meat quality traitsthe sensitivity of the values indicates that the selection must be made in the tryptophan:lysine ratio in which the quails will be created.

Conselho Nacional de Desenvolvimento Científico e Tecnológico / The study of adaptability and yield stability allows to identify cultivars adapted to specific conditions and wide, stable and with high grain yield, and can be realized as the different methodologies employed. Thus, the aim of this study was to investigate the association between adaptability and stability methods for grain yield in common bean cultivars and identify the most promising for cultivation in State of Rio Grande do Sul. For this, were used the Eberhart and Russell (1966), the Cruz et al. (1989), the additive main effects and multiplicative interaction analysis (AMMI), the Lin and Binns (1988) modified and the Annicchiarico (1992) methods, and, in the first three methods, the average grain yield was considered as a parameter of stability and adaptability. The competition trials of common bean cultivars were conducted in randomized block experimental design with three replications, during the growing seasons from 2000 to 2008, in ten different places of Rio Grande do Sul State, in different crop cultivation and harvesting period, totaling 24 experiments. We evaluated 12 common bean cultivars of different commercial groups (black, carioca and cranberry): Diamante Negro, Guapo Brilhante, Guateian 6662, IAPAR 44, Macanudo, Macotaço, Minuano, Tio Tibagi, TPS Nobre, Carioca, Pérola and Iraí. Significative cultivars x environments interaction were observed for grain yield. The TPS Nobre cultivar was the one closest to praised as ideal behavior, except for the AMMI analysis. While the Iraí cultivar was unstable and had low grain yield in Rio Grande do Sul State. Significative correlation was observed among Eberhart and Russell (1966), Cruz et al. (1989), AMMI, Lin and Binns (1988) modified and Annicchiarico (1992) methods, demonstrating that they are associated. Lin and Binns (1988) modified (rs = 0.95) and Annicchiarico (1992) (rs = 0.97) methodologies were highly correlated with the average grain yield and allowed the identification of stable and adapted cultivars among the most productive. Thus, in analysis of adaptability and stability for common bean crop, when the average grain yield is considered as a parameter of adaptability and stability, not recommended to use together of Eberhart and Russell (1966), Cruz et al. (1989), AMMI, Lin and Binns (1988) modified and Annicchiarico (1992) methods. Choosing the most appropriate method is at the discretion of the researcher. / O estudo da adaptabilidade e da estabilidade para produtividade de grãos permite a identificação de cultivares adaptadas às condições específicas ou amplas, estáveis e de elevada produtividade de grãos e pode ser realizado com o emprego de diferentes metodologias. Assim, o objetivo deste trabalho foi verificar a associação entre métodos de adaptabilidade e de estabilidade para a produtividade de grãos em feijão e identificar cultivares mais promissoras para o cultivo no Estado do Rio Grande do Sul. Para tanto, foram utilizados os métodos de Eberhart e Russell (1966), de Cruz et al. (1989), da análise dos efeitos principais aditivos e da interação multiplicativa (AMMI), de Lin e Binns (1988) modificado e de Annicchiarico (1992), sendo que, nas três primeiras metodologias, a média de produtividade de grãos foi considerada como um parâmetro de adaptabilidade e estabilidade. Os experimentos de competição de cultivares de feijão foram realizados no delineamento experimental de blocos ao acaso, com três repetições, durante os anos agrícolas de 2000 a 2008, em nove municípios do Estado do Rio Grande do Sul, em cultivo de safra e de safrinha, totalizando 24 experimentos. Foram avaliadas 12 cultivares de feijão de diferentes grupos comerciais (preto, carioca e manteigão): Diamante Negro, Guapo Brilhante, Guateian 6662, IAPAR 44, Macanudo, Macotaço, Minuano, Tio Tibagi, TPS Nobre, Carioca, Pérola e Iraí. Houve interação cultivares com ambientes significativa para a produtividade de grãos. A cultivar TPS Nobre foi a que mais se aproximou do comportamento preconizado como ideal, exceto pela análise AMMI. Já a cultivar Iraí foi instável e apresentou baixa produtividade de grãos no Estado do Rio Grande do Sul. Correlação significativa foi observada entres os métodos de Eberhart e Russell (1966), de Cruz et al. (1989), AMMI, de Lin e Binns (1988) modificado e de Annicchiarico (1992), demonstrando que estes estão associados. As metodologias de Lin e Binns (1988) modificado (rs = 0,95) e de Annicchiarico (1992) (rs = 0,97) foram altamente correlacionadas com as médias de produtividade de grãos e permitiram a identificação de cultivares estáveis e adaptadas entre as mais produtivas. Deste modo, em análise de adaptabilidade e de estabilidade na cultura do feijão, quando a média da produtividade de grãos é considerada como um parâmetro de adaptabilidade e de estabilidade, não se recomenda o uso conjunto dos métodos de Eberhart e Russell (1966), de Cruz et al. (1989), AMMI, de Lin e Binns (1988) modificado e de Annicchiarico (1992). A escolha do método mais adequado fica a critério do pesquisador.