Os objetivos deste trabalho foram estimar as associações genéticas entre características de crescimento e produção de leite, utilizando análises bicaracterísticas, e entre características de crescimento, perímetro escrotal e idade ao primeiro parto utilizando modelos de regressão aleatória, de animais da raça Guzerá. Para as análises bicaracterísticas foram utilizadas 252.257 informações de pesos de machos e fêmeas aos 120 dias (P120), ao desmame (PD), ao ano (P365), ao sobreano (PSOBRE) aos 24 meses de idade (P24) e 6.493 lactações encerradas (P305), pertencentes a 4.723 vacas, e os modelos incluíram como os efeitos aleatórios o genético aditivo direto, de ambiente permanente materno e residual, e como efeitos fixos os grupos de contemporâneos e a idade da vaca ao parto (efeito linear e quadrático). Para as análises com os modelos de regressão aleatória foram utilizadas 159.366 observações de pesos e 23.780 de perímetro escrotal, realizadas entre 335 e 724 dias de idade dos animais e agrupadas em classes com intervalo de 10 dias, e 63.596 observações de idade ao primeiro parto. Os efeitos aleatórios considerados foram o genético aditivo direto, de ambiente permanente e residual e como efeitos fixos os grupos de contemporâneos, a idade da vaca ao parto (efeito linear e quadrático) e a curva fixa para modelar a tendência média da população (quadrática) sobre as classes de idade. Foram testados quatro possíveis graus de polinômios de Legendre (zero, linear, quadrático e cúbico), sendo o quadrático mais adequado para descrever as variâncias das características analisadas. Para verificar a existência de diferentes padrões de crescimento e agrupar os touros com base nos seus valores genéticos para produção de carne, leite e duplo propósito, foram realizadas análises de componentes principais e de agrupamento. As herdabilidades estimadas foram 0,23; 0,14; 0,16; 0,17; 021 e 0,22 para P305, P120, PD, P365, PSOBRE e P24, respectivamente, sugerindo que para as características de pesos, as herdabilidades aumentam com o aumento da idade dos animais. Mesma tendência foi observada pelas análises de regressão aleatória, cujas herdabilidades variaram de 0,17 a 0,31. As correlações genéticas entre os pesos em diferentes idades e a produção de leite foram positivas e de magnitude moderada a baixa, variando de 0,27 a 0,38 sugerindo que a seleção para peso e P305 possa ser realizada de forma simultânea nos mesmos animais. As análises de componentes principais indicaram a mesma tendência observada pelas correlações genéticas. As análises de agrupamento mostraram que a raça Guzerá possui touros com diferentes perfis genéticos, sendo possível realizar a seleção para corte, leite ou duplo propósito. As correlações genéticas entre os pesos e perímetro escrotal foram positivas e favoráveis, variando de 0,31 a 0,47, indicando que a seleção para aumento do peso poderá resultar em animais com maior perímetro escrotal. As correlações entre peso e idade ao primeiro parto variaram de -0,56 a -0,38 e perímetro escrotal e idade ao primeiro parto variaram de -0,55 a 0,08, sugerindo que a redução da idade ao primeiro parto poderá ocorrer, a longo prazo, quando peso e perímetro escrotal forem objetivos de seleção. A eficiência relativa de seleção indicou maior resposta pela seleção indireta para idade ao primeiro parto, quando realizada a seleção para perímetro escrotal a partir dos 615 dias de idade, quando comparada com o ganho genético direto para idade ao primeiro parto. / The objectives of this study were to estimate the genetic associations between growth and milk production traits in Guzerat cattle using two-trait analysis and between growth traits, scrotal circumference and age at first calving using random regression models. For two-trait analysis, 252,257 weight records of males and females obtained at 120 days of age (W120), weaning (WW), yearling (YW), post-weaning (PWW) and 24 months of age (W24), as well as 6,493 complete lactation records (W305) of 4,723 cows, were used. The models included direct additive genetic, maternal permanent environmental and residual effects as random effects, and the contemporary groups and age of cow at calving (linear and quadratic effect) as fixed effects. For the random regression models, 159,366 observations of weight and 23,780 observations of scrotal circumference, obtained at 335 and 724 days of age of the animals and divided into classes at 10-day intervals, as well as 63,596 observations of age at first calving, were used. Random direct additive genetic, permanent environmental and residual effects and the fixed effects of contemporary group and age of cow at calving (linear and quadratic effect) were considered. A fixed curve was used to model the average trend of the population (quadratic) on the age classes. Four possible degrees of Legendre polynomials (zero, linear, quadratic and cubic) were tested and the quadratic polynomial was the most appropriate to describe the variances in the traits analyzed. Principal component and cluster analyses were performed to determine the existence of different growth patterns and to group bulls based on their breeding values for meat, milk and dual-purpose production. The estimated heritabilities were 0.23, 0.14, 0.16, 0.17, 0.21 and 0.22 for W305, W120, WW, YW, PWW and W24, respectively, suggesting that, for the growth traits, heritabilities increased with increasing age of the animals. The same trend was observed when random regression analysis was performed, with heritabilities ranging from 0.17 to 0.31. The genetic correlations between weights at different ages and milk yield were positive and of moderate to low magnitude, ranging from 0.27 to 0.38. These estimates suggest that selection for weight and W305 can be performed simultaneously in the same animals. Principal component analysis indicated the same trend as observed by the genetic correlations. Cluster analysis showed the presence of bulls with different genetic profiles in the Guzerat breed, thus permitting selection for meat, milk or dual purpose. The genetic correlations between weights and scrotal circumference were positive and favorable (0.31 to 0.47), indicating that selection for increased weight will result in animals with greater scrotal circumference. The correlations between weights and age at first calving ranged from -0.56 to -0.38 and between scrotal circumference and age at first calving from -0.55 to 0.08, suggesting that the use of weight and scrotal circumference as selection objectives will result in the long-term reduction of age at first calving. The relative efficiency of selection indicated a greater response to indirect selection for age at first calving when selecting for scrotal circumference after 615 days of age, compared to the genetic gain obtained by direct selection for age at first calving.

Orientadores: Jose Roberto Nunhez, Efraim Cekinski / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Quimica / Made available in DSpace on 2018-08-02T03:18:40Z (GMT). No. of bitstreams: 1 Silva_JeffersonLuizGrangeiroda_D.pdf: 4610700 bytes, checksum: b6db5ffbc0e16576244bd0eaea39201a (MD5) Previous issue date: 2002 / Resumo: O projeto atual de agitadores químicos é essencialmente dependente de relações empíricas, visto que as leis fisicas que regem os fenômenos envolvidos nos processos químicos são expressões cuja solução analítica é ainda desconhecida. Devido a isto, trabalhos experimentais como os compilados por Nagata, (1975), Oldshue, (1983), Sano e Usui, (1985) e Tatterson, (1991) são utilizados como principais referências para o cálculo do consumo de energia destes sistemas de agitação. Este projeto de pesquisa tem por objetivo principal compilar na literatura as principais correlações e trabalhos que são utilizados como base para o cálculo convencional de agitadores utilizados na indústria química, de forma que sejam definidas as melhores correlações através da análise dos parâmetros fundamentais (relações geométricas) no cálculo do Número de Potência, Número de Reynolds, Número de Froude, Número de Fluxo, Número de Mistura e o consumo de Potência de impelidores para os sistemas de agitação. A análise dos parâmetros geométricos será particularmente importante para estágios iniciais de projetos de plantas químicas, quando são necessárias informações da geometria, tanto do tanque como do impelidor (agitador) e, principalmente, do consumo de energia específica destes equipamentos para se estimar o consumo de energia da planta. Os resultados analisados através da utilização das correlações empíricas serão comparados com o Software comercial VISIMIX. Adicionalmente o L-CFD, laboratório de Fluído Dinâmica Computacional, já desenvolveu diversos projetos que analisam a otimização de projeto de reatores de mistura utilizando a Fluido Dinâmica Computacional e este projeto contribui por complementar o conhecimento deste grupo de pesquisa sobre o projeto de agitadores através de cálculos convencionais / Abstract: The design of impellers to date depends on empirical relations, since the physical laws that describe the phenomena involved in the chemical processes are expressions whose analytical solution is still unknown. Experimental works as those compiled by Nagata, (1975), Oldshue, (1983), Sano and Usui, (1985) and Tatterson, (1991) are used as references for energy consumption estimation in these systems. The compilation of the main correlations and the works in the literature that are used as a base to the conventional way to ca1culate the main impellers used in the chemical industry is the main objective of this research; which will allow one to select the better correlations it is carried on by the analysis of fundamental paramenters (geometric relations) for the ca1culation of Power Number, Reynolds Number, Froude Number, Flow Number, Mixing Number and the Power consumption of impellers in these systems. The analysis of the geometrical parameters will be particularly important to the early stages in the design of chemical plants, when informations ofthe tank and ofthe impeller (agitator) geometries and, mainly,about the specific energy consumption of these equipments are needed. The results are analysed using empirical correlations in the literature and are compared with the commercial software VISIMIX. Additionaly the L-CFD, Computational Fluid Dynamics laboratory, has already developed several projetcs to analyse the optimization project of mixing tanks, through the use of Computational Fluid Dynamics. This project contributes to complement the knowledge of this group of research on the design of mixing impellers, through the use of conventional methods / Doutorado / Desenvolvimento de Processos Químicos / Mestre em Engenharia Química

Le syndrome de Bardet-Biedl (BBS) est une ciliopathie syndromique associant une rétinopathie pigmentaire, une polydactylie post-axiale, une obésité, un hypogonadisme, des anomalies rénales et des troubles des apprentissages. Le cil primaire est présent à la surface de la quasi totalité des cellules de l’organisme et joue un rôle d’antenne cellulaire captant les signaux extérieurs pour les transmettre à la cellule. A ce jour 21 gènes BBS ont été identifiés codant des protéines ayant une fonction ciliaire. Au cours de ce travail, nous avons identifié 3 nouveaux gènes BBS (SDCCAG8/BBS16, LZTFL1/BBS17, BBIP1/BBS18) et confirmé l’implication de IFT172/BBS20. Nous avons également établi des corrélations génotype-phénotype : absence de polydactylie et insuffisance rénale associées aux mutations dans BBS16 ; polydactylie mésoaxiale et atteinte rénale associées aux mutations dans BBS17 ; possible association d’une polydactylie préaxiale aux mutations dans BBS20. Enfin, nous décrivons sur le plan clinique et moléculaire la plus grande cohorte de patients BBS à partir d’une base de données clinico-biologique mise en place au cours de ce travail. / Bardet-Biedl syndrome (BBS) is a syndromic ciliopathy associating with retinitis pigmentosa, postaxial polydactyly, obesity, hypogonadism, renal anomalies and learning difficulties. The primary cilium is antenna-like structure at the surface of the cell. 21 BBS genes are identified and the corresponding proteins are related to primary cilium structure and function. In this study, we identified 3 new BBS genes (SDCCAG8/BBS16, LZTFL1/BBS17, BBIP1/BBS18) and we confirmed the implication of IFT172/BBS20 in this syndrome. We also established strong genotype-phenotype correlations: absence of polydactyly and early renal failure in SDCCAG8/BBS16 patients; mesoaxial polydactyly and early renal failure in LZTFL1/BBS17 patients; possible preaxial polydactyly in IFT172/BBS20 patients. Finally, we reported the molecular and clinical description of the largest BBS cohort thanks to the clinical and biological database created in the Laboratory.

Les superalliages à base nickel des turbines aéronautiques sont susceptibles de subir des phénomènes de corrosion et/ou d’oxydation à haute température par les environnements agressifs rencontrés en service. Aussi, des revêtements d’aluminure sont appliqués par dépôt chimique afin d’assurer la protection des pièces contre ces phénomènes. La dégradation progressive de ces revêtements mène à la nécessité de les enlever afin d’en appliquer des nouveaux. Les bains chimiques industriels pour enlever les revêtements et les oxydes sont très toxiques, polluants et plutôt empiriques. Ainsi, ce travail de thèse se proposait d’étudier une méthode alternative et originale, par voie électrochimique permettant de contrer les limitations des approches chimiques. La voie électrochimique par application d’un potentiel (mode potentiostatique) a été étudiée afin de procurer la sélectivité entre le substrat et le revêtement lors de la dissolution, ainsi que le contrôle in-situ du procédé à l’aide d’une cellule à 3 électrodes. La faisabilité de la méthode a d’abord été démontrée, puis différentes procédures (par cycles cathodique/anodique, en continu et, parfois, avec modification du potentiel imposé) ont été développées. Nous avons pu mettre en relation les états métallurgiques des systèmes revêtement/substrat avec leur comportement électrochimique et avons mis également en lumière que le taux de dissolution est principalement gouverné par la concentration d’aluminium dans le revêtement alors que lorsque le platine est incorporé à ce même revêtement, le taux de dissolution est homogène. De même, nous avons démontré par XPS et par MET que la teneur en chrome modifie de manière significative l’homogénéité du décapage lors des phases de polarisation cathodique par un mécanisme de passivation de la surface, qui bloque l’activité électrochimique. Cependant, la tenue en oxydation cyclique des revêtements décapés par voies chimique et électrochimique n’a pas pu véritablement être démontrée car les revêtements avaient une microstructure différente. Enfin, des essais sur pièces de turbine ont montré le haut degré de sélectivité de l’approche ici étudiée. / Nickel based superalloys of aeronautical turbines are subjected to high temperature oxidation and/or corrosion in service conditions. Thus, protective aluminide coatings are applied onto the parts by chemical vapor deposition. The degradation of the coatings with time requires them to be removed prior to recoating the parts. The chemical baths industrially employed are toxic, polluting and quite empirical. Therefore, this thesis aimed at studying an alternative and original electrochemical method to circumvent the drawbacks of the chemical approach. Fixed potentials (potentiostatic mode) were thus applied to provide selectivity between the coating and the substrate upon the dissolution process, as well as to ensure in-situ control through a 3-electrode cell. The feasibility of the method was first demonstrated, then different procedures (cathodic/anodic cycles; continuous anodic and sometimes with modification of the potential) were investigated. The correlations between the metallurgical phases of the coating/substrate systems were elucidated. It also appeared that dissolution is mainly governed by the concentration of aluminium in the coating whereas the incorporation of platinum to the coating brought about the homogeneous dissolution. In addition, XPS and MET confirmed the hypothesis by which the chromium content drastically change the stripping homogeneity upon the cathodic polarization step by passivation of the surface and the subsequent electrochemical blocking. However, the results on the cyclic oxidation behaviour of the coatings priorly stripped chemically or electrochemically were not conclusive enough as the microstructure of the original coatings was different. Finally, quite a few stripping trials were carried out onto real turbine parts that confirmed the high selectivity of the electrochemical approach studied.

Anomalous properties of the normal state of a strongly correlated electron system described by an attractive extended Hubbard model are investigated. The equations of motion of the Green s functions are calculated with the two-pole approximation which gives rise to quasiparticle renormalized bands. The two-pole approximation leads to a set of correlation functions. In particular, the antiferromagnetic correlation function h~Si · ~Sji plays an important role as a source of anomalies in the normal state of the model. The uniform static magnetic susceptibility as a function of occupation nT and temperature is calculated. At low temperatures, the susceptibility presents a peak for nT ≃ 0.80. The results suggest that it is the onset of short range antiferromagnetic correlations, which could be a mechanism for the pseudogap. The Fermi surface, defined by the spectral function A(ω = 0,~k), is presented for different dopings. It has been observed that above nT ≃ 0.80 the ordinary Fermi surface evolves to a hole-pocket with pseudogaps near the antinodal points (0, π) and (π, 0). / Neste trabalho, investigamos certas propriedades anômalas do estado normal de sistemas de elétrons fortemente correlacionados, descrito por um modelo de Hubbard estendido, com interação atrativa. As equações de movimento das funções de Green são calculadas na aproximação de dois polos que gera às bandas de quasipartículas renormalizadas. A aproximação de dois polos dá origem a um conjunto de funções correlação. Em particular, a função correlação h~Si.~Sji, associadas ás correlações antiferromagnética, desempenha um papel importante como fonte de anomalias no estado normal do modelo. A susceptibilidade magnética é calculada como função da ocupação nT e da temperatura. Em baixas temperaturas, a susceptibilidade apresenta um pico para nT∼=0, 80 e é nessa ocupação que as correlações antiferromagnéticas assumem um papel importante responsável pelo surgimento de pseudogaps na superfície de Fermi. O cálculo do calor específico em função da temperatura mostra uma estrutura de dois picos, um associado ás flutuações de spin e localizado em baixas temperaturas e outro associado á flutuações de cargas localizado em temperaturas mais altas. Verificamos uma relação direta entre o pico, devido ás flutuações de spins e às correlações spin-spin do tipo antiferromagnéticas. A superfície de Fermi definida pela função espectral (A~k,σ(ω)) em ω = 0 é calculada para diferentes ocupações. Foi observado que a partir de nT∼=0, 80 a superfície de Fermi desenvolve pockets centrados no ponto nodal (π 2 , π 2 ) como também pseudogaps nas proximidades dos pontos antinodais (π, 0) e (0, π).

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The dual purpose wheat has the capacity to produce forage and grain, through the expression of morphological and bromatologic traits and yield components. Thus, the overall goal was to identify biometric models to understand the interrelationships between traits, direct and indirect selection in breeding wheat for dual purposes. The specific objectives of this study were to determine the canonical correlations between groups of morphological traits and grain yield, and identify the direction and the magnitude of the relationship; and identify the phenotypic relationships of cause and effect bromatologic between forage characters in wheat genotypes under different dual purpose cutting managements. The experiments were conducted in the growing seasons 2013 and 2014 in the experimental area of the Federal University of Santa Maria Campus Frederico Westphalen - RS. A randomized complete block design was used in three trials, the first experiment was arranged in a factorial: five genotypes (BRS Tarumã, BRS Umbu, BRS Figueira, BRS Guatambu and BRS 277) x four cutting managements (uncut, a cut, two cuts and three cuts). The second experiment was arranged in a factorial: Five genotypes (BRS Tarumã, BRS Umbu, BRS Figueira, BRS Guatambu and BRS 277) x Three cutting managements (first cut, second cut and third cut). The traits evaluated were: number of tillers per plant (AF), stem diameter of tillers (DAF), diameter of the main stem (DCP), number of fertile tillers (AFF), number of ears per square meter (ESPM), number of grains per ear (NGE), grain weight per ear (MGE), thousand grain weight (MMG), grain yield (RG), weight hectolitre (PH), green matter per hectare (MV), dry matter per hectare (MS), percentage of hemicellulose (HEM), cellulose (CEL), mineral matter (MM), neutral detergent fiber (FDN), acid detergent fiber (FDA), lignin (LIG), total carbohydrates (CT), carbohydrates no fiber carbohydrates (CNF), crude protein (PTN), and lipids (LIP). Groups of morphological traits and grain yield in wheat dual purpose reveal highly significant relationships. In the management at uncut and one cut wheat, associations are similarly established for both groups, by increasing the DCP, and determining the ESPM the increase, NGE and MGE, MMG, PH and RG. With two cuts the associations are established through the DAF, which influences an increase in MMG, PH and RG. Three cuts showed that the increase in AF increment the ESPM, NGE and MGE, MMG, PH and RG. The indirect selection aimed at increasing the PTN in the first cut can be based on CT, CNF and MV. The second cut provides indirect selection with LIG, CNF, MV and MS. The increase of PTN in the third cut can be obtained by indirect selection to the FDA, CEL, LIG, MM, MV and MS. Indirect selection through forage characters and bromatologic can be successful, since it considers the effect assigned to each cutting management. / O trigo duplo propósito apresenta capacidade de produzir forragem e grãos, através da expressão de caracteres morfológicos, bromatológicos e componentes do rendimento. Desta maneira, o objetivo geral foi identificar modelos biométricos que permitam compreender as inter-relações entre caracteres, e direcionar a seleção indireta no melhoramento genético de trigo duplo propósito. Os objetivos específicos deste trabalho foram: determinar as correlações canônicas entre grupos de caracteres morfológicos e do rendimento de grãos, e identificar o sentido e a magnitude das relações; e identificar as associações fenotípicas de causa e efeito entre caracteres forrageiros e bromatológicos em genótipos de trigo duplo propósito submetidos a diferentes manejos de corte.Os experimentos foram realizados nas safras agrícolas 2013 e 2014 na área experimental da Universidade Federal de Santa Maria Campus de Frederico Westphalen RS. Utilizou-se o delineamento de blocos ao acaso em três repetições, o primeiro experimento foi organizado em fatorial: Cinco genótipos (BRS Tarumã, BRS Umbu, BRS Figueira, BRS Guatambu e BRS 277) x Quatro manejos de corte (sem corte, um corte, dois cortes e três cortes). O segundo experimento foi organizado em fatorial: Cinco genótipos (BRS Tarumã, BRS Umbu, BRS Figueira, BRS Guatambu e BRS 277) x Três manejos de corte (primeiro corte, segundo corte e terceiro corte). Os caracteres avaliados foram: número de afilhos por planta (AF), diâmetro do colmo dos afilhos (DAF), diâmetro do colmo principal (DCP), número de afilhos férteis (AFF), número de espigas por metro quadrado (ESPM), número de grãos por espiga (NGE), massa de grãos por espiga (MGE), massa de mil grãos (MMG), rendimento de grãos (RG), peso hectolitro (PH), massa verde por hectare (MV), massa seca por hectare (MS), percentual de hemicelulose (HEM), celulose (CEL), material mineral (MM), fibras em detergente neutro (FDN), fibras em detergente ácido (FDA), lignina (LIG), carboidratos totais (CT), carboidratos não fibrosos (CNF), proteína bruta (PTN), e lipídios (LIP). Os grupos de caracteres morfológicos e de rendimento de grãos em trigo duplo propósito revelam relações altamente significativas. No manejo sem cortes e com um corte, associações são estabelecidas similarmente para ambos os grupos, através do aumento do DCP, sendo determinante ao acréscimo de ESPM, NGE e MGE, MMG, PH e RG. Com dois cortes as associações são estabelecidas através do DAF, que influencia o aumento da MMG, PH e RG. Com três cortes evidencia-se que o aumento do AF incrementam ESPM,NGE e MGE, MMG, PH e RG.A seleção indireta visando incrementar PTN no primeiro corte pode ser baseada nos CT,CNF e MV. O segundo corte proporciona a seleção indireta com a LIG, CNF, MV e MS. O incremento de PTN no terceiro corte poderá ser obtido através da seleção indireta com FDA, CEL, LIG, MM, MV e MS. A seleção indireta através dos caracteres forrageiros e bromatológicos pode ser realizada com sucesso, desde que se considere o efeito atribuído a cada manejo de corte.

Le déficit en 21-hydroxylase est la cause la plus fréquente des hyperplasies congénitales des surrénales. Un grand nombre de nouvelles mutations a été trouvé dans le laboratoire qui centralise la plus grande cohorte de familles au niveau international et l’évaluation de leur sévérité était primordiale pour optimiser la prise en charge des patients (thérapeutique, conseil génétique). Grâce à l’analyse approfondie du phénotype des patients et au développement d’études fonctionnelles (in vitro, in silico), nous avons pu évaluer le retentissement de la plupart des 85 nouvelles mutations ; nous avons choisi comme témoins des mutations fréquentes de sévérité connue et nous avons comparé nos résultats avec ceux de la littérature. L’analyse plus approfondie d’une quinzaine de mutations rares a confirmé l’existence de bonnes corrélations phénotype-génotype comme ceci est décrit dans cette pathologie. Par ailleurs, les études structurales que nous avons développées ont permis d’améliorer les connaissances sur les relations structure-fonction des cytochromes P450 en général. / Steroid 21-hydroxylase deficiency is the most common enzymatic defect causing congenita ladrenal hyperplasia. A large number of new mutations has been detected in the laboratory, which centralizes the biggest cohort of families in the world, and evaluation of their severity wasessential to optimize the care of the patients (treatment, genetic counselling). Thanks to detailed analysis of the patients phenotype and to the development of functional studies (in vitro, in silico), we were able to evaluate the severity of most of the 85 novel mutations; we decided touse as controls frequent known mutations and to compare our results with those of literature. Themore detailed analysis of about fifteen rare mutations confirmed the existence of goodcorrelations phenotype-genotype as this is described in this pathology. Moreover, the structural studies we developed led to improve the knowledge on structure-function relationship of theP450 cytochromes family.

The amount of digitally-born data has surged in recent years. In many scenarios, this data is inherently uncertain (or: probabilistic), such as data originating from sensor networks, image and voice recognition, location detection, and automated web data extraction. Probabilistic data requires novel and different approaches to data mining and analysis, which explicitly account for the uncertainty and the correlations therein. This thesis introduces ENFrame: a framework for processing and mining correlated probabilistic data. Using this framework, it is possible to express both traditional and novel algorithms for data analysis in a special user language, without having to explicitly address the uncertainty of the data on which the algorithms operate. The framework will subsequently execute the algorithm on the probabilistic input, and perform exact or approximate parallel probability computation. During the probability computation, correlations and provenance are succinctly encoded using probabilistic events. This thesis contains novel contributions in several directions. An expressive user language – a subset of Python – is introduced, which allows a programmer to implement algorithms for probabilistic data without requiring knowledge of the underlying probabilistic model. Furthermore, an event language is presented, which is used for the probabilistic interpretation of the user program. The event language can succinctly encode arbitrary correlations using events, which are the probabilistic counterparts of deterministic user program variables. These highly interconnected events are stored in an event network, a probabilistic interpretation of the original user program. Multiple techniques for exact and approximate probability computation (with error guarantees) of such event networks are presented, as well as techniques for parallel computation. Adaptations of multiple existing data mining algorithms are shown to work in the framework, and are subsequently subjected to an extensive experimental evaluation. Additionally, a use-case is presented in which a probabilistic adaptation of a clustering algorithm is used to predict faults in energy distribution networks. Lastly, this thesis presents techniques for integrating a number of different probabilistic data formalisms for use in this framework and in other applications.

Magister Psychologiae - MPsych / HIV and AIDS are disease conditions that have led to high mortality rates in Southern Africa since the late 1970s. The socio-economic system has led to the unequal spread of resources’ and vulnerability and exposure to HIV is more prevalent in poorer communities. The added burden of life stresses cause for many to be isolated and stigmatised and are often not equipped with the necessary support and coping skills to deal with the magnitude of these circumstances. There is a high prevalence of mental disorders and especially depression amongst individuals infected with either HIV or AIDS. Research shows that stressful life events can impact HIV course progression and impacts the QoL of those infected with HIV or AIDS. Given the social and psychological context of HIV and AIDS, the aim of the study was to examine the relationship between depression and QoL in a sample of adults diagnosed with HIV or AIDS. This quantitative, cross-sectional study used the Becks Depression Inventory II (BDI II) and the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q), to measure the variables concerned. This battery of questionnaires was administered to a purposive sample of adult individuals diagnosed with HIV or AIDS residing in a previously disadvantaged area in the Cape Metropole region. Using SPSS, data was analysed and descriptive and inferential statistics were conducted. The study found that there were more women than men with HIV or AIDS that were found to be depressed (mild, moderate and severe depression). Furthermore, the depressive state increased when the progression of the disease increased. There were generally no significant differences in the QoL experienced within various areas of life and overall life satisfaction experienced. However, the QoL experienced in work was lower. There was a significant relationship between the depressed state and QoL and life satisfaction experienced in household duties and tasks. The contribution of this study includes informing the larger research project, with regards to future treatment regimes. It will update statistics on the prevalence of depression and QoL of adults diagnosed with HIV or AIDS in the area. This study is framed within a biopsychosocial model and is theoretically underpinned by Beck’s theory of depression. Key words: HIV, AIDS, adults, depression, quality of life (QoL), Beck Depression Inventory II (BDI II), Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q), Antiretrovirals (ARV’s), prevalence data, correlations. / South Africa

My dissertation investigates the financial linkages and transmission of economic shocks between the US and the smallest emerging markets (frontier markets). The first chapter sets up an empirical model that examines the impact of US market returns and conditional volatility on the returns and conditional volatilities of twenty-one frontier markets. The model is estimated via maximum likelihood; utilizes the GARCH model of errors, and is applied to daily country data from the MSCI Barra. We find limited, but statistically significant exposure of Frontier markets to shocks from the US. Our results suggest that it is not the lagged US market returns that have impact; rather it is the expected US market returns that influence frontier market returns The second chapter sets up an empirical time-varying parameter (TVP) model to explore the time-variation in the impact of mean US returns on mean Frontier market returns. The model utilizes the Kalman filter algorithm as well as the GARCH model of errors and is applied to daily country data from the MSCI Barra. The TVP model detects statistically significant time-variation in the impact of US returns and low, but statistically and quantitatively important impact of US market conditional volatility. The third chapter studies the risk-return relationship in twenty Frontier country stock markets by setting up an international version of the intertemporal capital asset pricing model. The systematic risk in this model comes from covariance of Frontier market stock index returns with world returns. Both the systematic risk and risk premium are time-varying in our model. We also incorporate own country variances as additional determinants of Frontier country returns. Our results suggest statistically significant impact of both world and own country risk in explaining Frontier country returns. Time-variation in the world risk premium is also found to be statistically significant for most Frontier market returns. However, own country risk is found to be quantitatively more important.