O restabelecimento de energia em sistemas de distribuição de energia elétrica radiais geralmente envolve a reconfiguração de redes para restaurar eletricidade à(s) área(s) fora de serviço. As principais técnicas para restabelecimento de energia em sistemas de distribuição de grande porte têm sido os algoritmos evolutivos (AEs). Após a falta ter sido identificada e a zona em falta ter sido isolada do sistema, o algoritmo deve encontrar soluções em que: 1) supra com energia o maior número de consumidores possível, 2) minimize o número de operações de chaveamentos, 3) não viole restrições operacionais do sistema, 4) reduza o total de perdas resistivas, 5) a configuração da rede seja radial e, 6) obtenha tal solução em tempo real. Este projeto emprega uma nova estrutura de dados para manipular grafos produzindo exclusivamente configurações radiais e conexas, chamada representação nó-profundidade (RNP), garantindo que todas as soluções potenciais geradas pelo algoritmo satisfaçam os itens (1) e (5). Além disso, propõe-se um AE utilizando a RNP capaz de encontrar planos de restabelecimento adequados para sistemas de distribuição de larga-escala, com milhares de chaves e barras, em tempo real. / Energy restoration in radial distribution systems usually involves the network reconfiguration to restore the electricity to the out-of-service areas. The main approaches for energy restoration in large-scale distribution systems have been the evolutionary algorithms (EAs). After a fault has been identified and isolated, the algorithm must find solutions that: 1) supply energy to the larger number of consumers, 2) reduce the number of switching operations, 3) respect operational constraints of the system, 4) reduce the amount of power losses, 5) generate exclusively radial configurations and 6) find solutions in real time. This work uses a new data structure, called node-depth encoding (NDE), to manipulate graphs producing exclusively radial and connected configurations, and guaranteeing that all potential solutions generated by the algorithm satisfy items (1) and (5). Moreover, we propose an EA using the NDE that is capable of finding adequate restoration plans in real time for large-scale distribution systems, with thousands of switches and buses.

A introdução de software em sistemas de aplicações críticas traz consigo questões relacionadas à segurança (safety) que, durante muito tempo recaíram predominantemente sobre o desenvolvimento do hardware que compunha tais sistemas. Atualmente, padrões relacionados à segurança de software avaliam qualitativamente o impacto do seu uso sobre sistemas suscetíveis a falhas de natureza randômica. A pesquisa aqui desenvolvida visa, em complemento a outras investigações já realizadas, avaliar quantitativamente diversas técnicas de programação defensiva em função de sua representatividade no quesito segurança de sistemas de aplicação crítica tolerantes a erros. Como objetivo essencial, buscou-se avaliar o comportamento adquirido por um sistema tolerante a erros quando submetido a um processo de injeção de falhas por software. A tolerância a erros do sistema de aplicação crítica em estudo é alcançada, através de técnicas de programação defensiva aplicadas ao software original. Foram aplicadas diversas técnicas de programação defensiva e diversas combinações entre elas, de modo que foi possível avaliar quantitativamente e identificar possíveis padrões de níveis de segurança adquiridos em cada caso. / The introduction of software systems for critical applications raises safety issues that have long fell predominantly on the development of the hardware composing such systems. Currently, standards related to safety software qualitatively assess the impact of their use on systems sensitive to random errors. The research developed here seeks, in addition to other previous investigations, to quantitatively evaluate different techniques of defensive programming in function of their safety level in fault-tolerant safety critical systems. As a key objective, we sought to evaluate the behavior acquired by a fault-tolerant system when subjected to a software fault injection process. The fault-tolerance system, in a typical critical application under study, is achieved through the application of defensive programming techniques over the original software. Many defensive programming techniques and various combinations among them were applied, hence making it possible to quantitatively assess and identify possible patterns of safety levels acquired in each case.

Desenvolver um sistema de distribuição de energia confiável é certamente um desafio inerente aos profissionais do setor elétrico. Porém, os sistemas estão sujeitos a falhas e, sendo assim, o rápido restabelecimento traz a satisfação do cliente e reduz as compensações pagas pelas concessionárias de energia. Neste contexto a presente dissertação visa avaliar a metodologia para restabelecimento de energia denominada AEMT-H que se baseia em algoritmos evolutivos multi-objetivo se na estrutura de dados chamada Representação Nó- Profundidade (RNP). Esta avaliação ocorrerá através da aplicação do AEMT-H para obtenção de planos de restabelecimento de energia após a ocorrência de faltas simples no sistema COPEL da cidade de Londrina. Os resultados gerados serão avaliados estatisticamente e ainda subjetivamente pelos profissionais do COD da concessionária. Os algoritmos evolutivos têm apresentado resultados animadores para os problemas de restabelecimento de energia. Em especial, os resultados obtidos, quando da representação computacional de sistemas de distribuição de grande porte (com milhares de barras e chaves) através da RNP, possuem como característica o tempo de resposta da ordem de segundos, instigando a evolução das pesquisas para utilização desta metodologia inclusive em aplicativos de tempo real. O sistema de Londrina possui 30.156 barras, 2.660 chaves \"NF\", 250 chaves \"NA\" e atende um universo de mais de 231.000 consumidores ligados diretamente ao sistema de 13,8 kV ou, após os transformadores de distribuição, nas tensões de 220V e 127V. Neste sistema estão presentes 6 subestações 138 kV/13,8 kV e 64 circuitos alimentadores, totalizando uma capacidade de transformação de energia de 541,7 MVA. / The development of a reliable distribution system is certainly a challenge to electrical industry professionals. However, these systems are subject to failures and thus the fast restoration brings customer satisfaction and reduces the compensation paid by the electricity utilities. In this context, this dissertation aims to evaluate the methodology for service restoration called \"AEMT-H\", which is based on multi-objective evolutionary algorithms and in the data structure called Node Depth Encoding (NDE). This evaluation will occur by applying the \"AEMT-H\" to obtain service restoration plans considering the occurrence of simple faults in the COPEL system of the city of Londrina. The obtained results will bee valuated statistically and subjectively by professionals of the Distribution Operation Center. It is important to highlight that Evolutionary algorithms have shown promising results to treat the service restoration problem in distribution systems. In particular, the results obtained when using NDE tocomputationally represent the electrical topology of large distribution systems (with thousand of buses and switchers) are very interesting in terms of time processing (in the order of seconds). The system of Londrina has 30,156 buses, 2,660 switchers normally closed, 250 switchers normally opened and supplies a universe of more than 231,000 consumers connected directly to the system with voltage of 13.8 kV or, after distribution transformers, with voltages of 220V and 127V. This system contains 6 substations 138 kV/13,8 kV and 64 feeders, and a installed power of 541.7 MVA.

Submitted by SBI Biblioteca Digital (sbi.bibliotecadigital@puc-campinas.edu.br) on 2016-05-10T19:33:58Z No. of bitstreams: 1 Karla Oliveira Franco.pdf: 1221586 bytes, checksum: 0c45f3f2ae3f0f3cf0e12dd46f9383d6 (MD5) / Made available in DSpace on 2016-05-10T19:33:58Z (GMT). No. of bitstreams: 1 Karla Oliveira Franco.pdf: 1221586 bytes, checksum: 0c45f3f2ae3f0f3cf0e12dd46f9383d6 (MD5) Previous issue date: 2016-02-25 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES / This master?s dissertation is inserted in the scope of the studies produced on assessment state systems on large-scale, and specifically researched the Sistema Mineiro de Avalia??o da Educa??o P?blica (SIMAVE) through a qualitative study developed from bibliographic search. The objective was to conduct a study about the State of The Art directed to SIMAVE, focusing on the identification of the respective evaluation system impacts in the school units from Minas Gerais, whose object of study was the masters and doctoral dissertations produced in Brazil during the years 2000 to 2013. The specific objectives was designed to study the path of the evaluation system of the implementation process on a large-scale for Basic Education in the State of Minas Gerais, in order to identify the key steps in the consolidation process of SIMAVE; the mapping of the scientific production, expressed through dissertations and doctoral theses produced since the implementation of SIMAVE in 2000-2013; and, the identification of the impacts of SIMAVE on the school units from Minas Gerais. To describe the trajectory of SIMAVE, three stages were identified in the process of consolidation, and verified that they go to three generations of large-scale assessments of Basic Education of Brazil, proposed by Bonamino and Sousa (2012), which are differentiated from the consequences attached to schools and the school curriculum. The systematization of scientific production resulted in 33 studies, in which 28 were dissertations and 5 doctoral theses, which were developed, mostly in post-graduate programs in Education. Through the research problem, which was to verify what were the impacts caused by SIMAVE at schools of Minas Gerais, it concludes that the SIMAVE caused impacts on Minas Gerais school units, and these were characterized from effects, weaknesses, risks and strategies. The identified impacts were subject to valuation both positive and negative, trough a confrontation of ideas in the context of the consensus paradigm and the conflict paradigm, reaffirming polysemy of the evaluation area, which is characterized by political and ideological tensions that were evident in the studies analyzed in this dissertation. / A presente disserta??o de mestrado enquadra-se no ?mbito dos estudos produzidos sobre os sistemas estaduais de avalia??o em larga escala, e tratou, especificamente, sobre o Sistema Mineiro de Avalia??o da Educa??o P?blica (SIMAVE) por meio de abordagem qualitativa desenvolvida a partir de pesquisa bibliogr?fica. O objetivo foi realizar um estudo do tipo Estado da Arte em torno do SIMAVE, com foco na identifica??o dos impactos do respectivo sistema de avalia??o nas unidades escolares mineiras, tendo como objeto de estudo disserta??es de mestrado e teses de doutorado produzidas no Brasil, no per?odo de 2000 a 2013. Estudos do tipo Estado da Arte permitem visualizar o que vem sendo produzido em determinada ?rea do conhecimento e realizar uma ordena??o que possibilita a percep??o da evolu??o cient?fica, bem como delimitar suas caracter?sticas e foco, al?m de possibilitar a identifica??o de lacunas existentes por meio de levantamento, an?lise e avalia??o de produ??es cient?ficas. Como objetivos espec?ficos desta pesquisa foi delineado o estudo da trajet?ria do processo de implanta??o do sistema de avalia??o em larga escala para Educa??o B?sica no Estado de Minas Gerais, visando identificar as principais etapas existentes no processo de consolida??o do SIMAVE; o mapeamento da produ??o cient?fica brasileira, expressa por meio de disserta??es de mestrado e teses de doutorado; e, a identifica??o dos impactos do SIMAVE nas unidades escolares mineiras. Ao tra?ar a trajet?ria do SIMAVE foram identificadas tr?s etapas no seu processo de consolida??o, e verificado que o mesmo se articula com as tr?s gera??es de avalia??es em larga escala da Educa??o B?sica do Brasil, propostas por Bonamino e Sousa, as quais s?o diferenciadas a partir das consequ?ncias atribu?das para as escolas e para o curr?culo escolar. Durante a pesquisa bibliogr?fica foram identificadas 33 produ??es acad?micas sobre o SIMAVE, sendo 28 disserta??es de mestrado e cinco teses de doutorado, que foram desenvolvidas, em sua maioria, em programas de p?s-gradua??o em Educa??o. Diante do problema de pesquisa, o qual consistiu em verificar quais foram os impactos provocados pelo SIMAVE nas unidades escolares de MG, conclui-se que o SIMAVE causou impactos nas unidades escolares mineiras, e estes foram caracterizados a partir de efeitos, fragilidades, riscos e estrat?gias de melhorias desenvolvidas em prol de melhor rendimento dos estudantes nos testes avaliativos do SIMAVE. Os impactos identificados foram pass?veis de valora??o tanto positiva quanto negativa, a partir de um confronto de ideias no contexto do paradigma do consenso e do paradigma do conflito, reafirmando a polissemia do campo da avalia??o, o qual ? caracterizado por tens?es pol?ticas e ideol?gicas que ficaram evidentes nos estudos analisados nesta disserta??o.

The large marine protected area (MPA) declared in 2010 around the Chagos Archipelago, also known as the British Indian Ocean Territory (BIOT), has led to a conflict in the thick of environmental protection, colonialism, sovereignty claims, and the human rights dispute of the Chagossian people that once were exiled from the islands in the 1970s. By applying a Foucauldian inspired critical discourse analysis, this study interprets and examines how the nature/human relationship was portrayed during and after the creation of the Chagos MPA. Applying theories and concepts from political ecology and Foucault’s idea of biopower sheds new light on a conservation effort depicted as a global environmental success by some, and a geopolitical social justice disaster by others. Finally, this thesis applies Tim Ingold’s philosophical concept of the globe and sphere to discuss the implications of inclusion or withdrawal from nature.

Audiobook reading has traditionally been used for educational purposes but has in recent times grown into a popular alternative to the more traditional means of consuming literature. In order to differentiate themselves from other players in the market, but also provide their users enjoyable literature, several audiobook companies have lately directed their efforts on producing own content. Creating highly rated content is, however, no easy task and one reoccurring challenge is how to make a bestselling story. In an attempt to identify latent features shared by successful audiobooks and evaluate proposed methods for literary quantification, this thesis employs an array of frameworks from the field of Statistics, Machine Learning and Natural Language Processing on data and literature provided by Storytel - Sweden’s largest audiobook company. We analyze and identify important features from a collection of 3077 Swedish books concerning their promotional and literary success. By considering features from the aspects Metadata, Theme, Plot, Style and Readability, we found that popular books are typically published as a book series, cover 1-3 central topics, write about, e.g., daughter-mother relationships and human closeness but that they also hold, on average, a higher proportion of verbs and a lower degree of short words. Despite successfully identifying these, but also other factors, we recognized that none of our models predicted “bestseller” adequately and that future work may desire to study additional factors, employ other models or even use different metrics to define and measure popularity. From our evaluation of the literary quantification methods, namely topic modeling and narrative approximation, we found that these methods are, in general, suitable for Swedish texts but that they require further improvement and experimentation to be successfully deployed for Swedish literature. For topic modeling, we recognized that the sole use of nouns provided more interpretable topics and that the inclusion of character names tended to pollute the topics. We also identified and discussed the possible problem of word inflections when modeling topics for more morphologically complex languages, and that additional preprocessing treatments such as word lemmatization or post-training text normalization may improve the quality and interpretability of topics. For the narrative approximation, we discovered that the method currently suffers from three shortcomings: (1) unreliable sentence segmentation, (2) unsatisfactory dictionary-based sentiment analysis and (3) the possible loss of sentiment information induced by translations. Despite only examining a handful of literary work, we further found that books written initially in Swedish had narratives that were more cross-language consistent compared to books written in English and then translated to Swedish.

Submitted by Rosina Valeria Lanzellotti Mattiussi Teixeira (rosina.teixeira@unisantos.br) on 2017-02-21T18:23:21Z No. of bitstreams: 1 Karla A. dos Reis Mehanna.pdf: 23364770 bytes, checksum: b752a9829dff02e2a50dfcaf3975177c (MD5) / Made available in DSpace on 2017-02-21T18:23:21Z (GMT). No. of bitstreams: 1 Karla A. dos Reis Mehanna.pdf: 23364770 bytes, checksum: b752a9829dff02e2a50dfcaf3975177c (MD5) Previous issue date: 2016-12-13 / The present dissertation presents results of a research that analyzed the uses of the ¿Prova Brasil¿ in two municipal schools of a metropolitan region, with the objectives of investigating its utilization for the school planning and to analyze if the continued formation of the teachers in the school, in the Schedule of Pedagogical Work (HTP), contemplates the study of this instrument of evaluation in large scale, the analysis of the results of the ¿Prova Brasil¿ and the proposition of actions to remedy any difficulties found by the school or improve the quality of work there. The research adopts a qualitative approach, involves two public primary schools (1st to 5th year), its school administrators and teachers. It adopts as methodological procedures: bibliographical survey and elaboration of reference on the subject; analysis of official and school documents; and semi-structured interviews. The findings of the bibliographical survey point to the diversification of the uses of the ¿Prova Brasil¿, among them: continued formation of the teachers; encouraging the creation of own systems for evaluation of networks and assistance in the selection of contents; policies of teacher accountability for student achievement; loss of school autonomy and curriculum reduction; pedagogical actions that only aim at the preparation for the tests; elaboration of rankings; and, even, the inexistence of the use of the ¿Prova Brasil¿ in schools. The findings in the documents of the studied schools indicate that there was no specific training work on the ¿Prova Brasil¿ during the year 2015 and that the time allocated to the training of teachers in the school is divided with questions of school organization, which occupy more space in the difficult to deepen the issues surrounding the evaluation. However, it was verified the use of the ¿Prova Brasil¿ as an evaluation model by the teachers and, for a plot, as a reference for self-evaluation of the work done in the classroom. Based on the constructed benchmark, it¿s considered that ¿Prova Brasil¿ permeates school planning, but its reports are not observed for the elaboration of the Political Pedagogical Project. As for continuing education in the school, the analysis of the minutes of meetings of the HTPs indicates that this activity needs to be rethought, since it does not fulfill its formative role. Teachers and managers consider the analysis of ¿Prova Brasil¿ and its results important, but they understand that this work becomes impracticable due to the short time available for this purpose. Considering the results, it¿s concluded that if the teacher training time in the school was better optimized, it would be possible to use the ¿Prova Brasil¿ as a subsidy in the school planning process and the continuous training of teachers in the school, due to the positive receptivity of this external evaluation instrument in schools, both by managers and teachers. / A presente dissertação apresenta resultados de uma pesquisa que analisou os usos da Prova Brasil por duas escolas municipais de uma região metropolitana, com os objetivos de investigar o seu emprego para o planejamento escolar e analisar se a formação continuada dos professores na escola, no Horário de Trabalho Pedagógico (HTP), contempla o estudo desse instrumento de avaliação em larga escala, a análise dos resultados da Prova Brasil e a proposição de ações para sanar eventuais dificuldades encontradas pela escola ou melhorar a qualidade do trabalho ali desenvolvido. A pesquisa adota abordagem qualitativa, envolvendo duas escolas públicas de Ensino Fundamental I (1º ao 5º ano), com seus gestores escolares e professores. Como procedimentos metodológicos, foram empregados: levantamento bibliográfico e elaboração de referencial sobre o tema; análise de documentos oficiais e escolares; e entrevistas semiestruturadas. Os achados do levantamento bibliográfico apontam para a diversificação dos usos da Prova Brasil, dentre eles: formação continuada dos professores; incentivo à criação de sistemas próprios de avaliação das redes e auxílio na seleção de conteúdos; políticas de responsabilização dos professores pelo rendimento dos alunos; perda de autonomia da escola e redução do currículo; ações pedagógicas que visam somente à preparação para os testes; elaboração de rankings; e, inclusive, inexistência do uso da Prova Brasil em escolas. Os achados nos documentos das escolas estudadas indicam que não houve trabalho de formação específico sobre a Prova Brasil durante o ano de 2015 e que o horário destinado à formação dos professores na escola é dividido com questões de organização escolar, as quais ocupam maior espaço nas reuniões, dificultando o aprofundamento dos assuntos que envolvem a avaliação. Entretanto, foi verificado o uso da Prova Brasil como modelo de avaliação pelos professores e, para uma parcela, como referência para autoavaliação do trabalho realizado em sala de aula. A partir do referencial construído, considera-se que a Prova Brasil permeia o planejamento escolar, mas seus relatórios não são observados para a elaboração do Projeto Político-Pedagógico. Quanto à formação continuada na escola, a análise das atas de reuniões dos HTP¿s indica que essa atividade precisa ser repensada, pois não cumpre seu papel formador. Professores e gestores consideram importante a análise da Prova Brasil e de seus resultados, mas entendem que esse trabalho se torna inviável em virtude do pouco tempo disponibilizado para tal finalidade. Diante dos resultados, conclui-se que, se o tempo de formação do professor na escola fosse mais bem otimizado, seria possível a utilização da Prova Brasil como subsídio no processo de planejamento escolar e de formação continuada de professores na escola, considerando a receptividade positiva desse instrumento de avaliação externa nas escolas, tanto por gestores quanto por professores.

Encontrar a configuração de mínimas perdas ôhmicas ou que forneça um adequado plano de restabelecimento aptos a ocorrência de faltas permanentes são problemas de natureza combinatorial, com múltiplos objetivos e restrições, que envolvem funções cujas características, em geral, dificultam o uso de técnicas de programação matemática. Algoritmos Evolutivos têm apresentado resultados animadores para esses problemas, especialmente quando aplicados em sistemas de distribuição de grande porte. Neste trabalho são propostas duas novas metodologias, baseadas em Algoritmos Evolutivos Multi-Objetivo, para reconfiguração de redes em sistemas de distribuição, que podem ser aplicadas para o tratamento dos problemas de redução de perdas ôhmicas e restabelecimento de energia em sistemas de grande porte (com milhares de barras, linhas e chaves). Para o desenvolvimento dessas metodologias, foi utilizado uma estrutura de dados para manipular grafos produzindo exclusivamente configurações radiais e conexas, chamada Representação Nó-profundidade. As metodologias propostas foram testadas para diversos sistemas, dentre os quais destaca-se um com 30:880 barras, tendo sido satisfatórios os resultados obtidos. Para o problema de restabelecimento de energia foram testados falta única e múltiplas faltas. / Find the distribution system configuration of minimum power losses or that provides an adequate service restoration plan is a combinatorial, multi-objective and multi constraint problems, which involves functions whose characteristics, in general, difficult the use of mathematical programming techniques. Evolutionary Algorithms have shown relevant results for these problems, especially for Large-Scale Distribution Systems. This work proposes two methodologies for network reconfiguration based on Multi-Objective Evolutionary Algorithms, which can be applied to treat the problems of power loss reduction and service restoration in large scale distribution systems (with thousands of buses, lines and switches). In order to develop these methodolgies, it was used a data structure to manipulate graphs producing exclusively radial and connected connections, called Node Depth Encoding. The proposed methodologies were successfully tested in several distribution systems, among them one with 30,880 buses. The problem of service restoration is analyzed considering cases of single and multiple faults.

O diabetes mellitus tipo MODY (maturity-onset diabetes of the young) é caracterizado por defeito na secreção de insulina, herança autossômica dominante, hiperglicemia de início precoce e anticorpos anti-células beta negativos. Até o momento, já foram descritas mutações em 14 genes diferentes. A confirmação do diagnóstico de MODY é feita por estudo genético-molecular, tradicionalmente pelo método de Sanger. Diante da grande heterogeneidade genética de MODY, acrescida da dificuldade de estudo de alguns genes por seu grande tamanho e ausência de hotspots, o sequenciamento em larga escala (SLE) mostra-se promissor para uma análise genética custo-efetiva na suspeita de MODY. No Brasil, existem poucos estudos genéticos de rastreamento de MODY e uma alta prevalência de casos sem mutações identificadas nos genes testados (MODY X). Os objetivos deste estudo foram: 1) analisar simultaneamente todos os genes associados a MODY em uma coorte de pacientes com suspeita clínica, utilizando um painel de SLE; 2) avaliar a patogenicidade das variantes alélicas identificadas de acordo com os critérios da Sociedade Americana de Genética Médica (ACMG). Foram selecionados 80 casos com fenótipo de MODY e análise prévia negativa dos 2 genes mais prevalentes, GCK e HNF1A, pelo método de sequenciamento de Sanger. Estes casos foram analisados pelo método de SLE, direcionado para regiões gênicas alvo, por meio de um painel customizado, com sequenciamento simultâneo de 51 genes nucleares e do genoma mitocondrial. As mutações identificadas foram correlacionadas com o fenótipo e foi realizada a segregação familiar. Uma cobertura de no mínimo 20x foi obtida em 98% das regiões alvo. Dos 80 pacientes avaliados, foram detectadas variantes patogênicas/potencialmente patogênicas em 16 casos (20%), confirmando o diagnóstico genético de MODY. Em 15 dos 80 pacientes foram identificadas 16 variantes de significado incerto, restando ainda 42 casos com diagnóstico molecular não esclarecido. Dos 16 casos confirmados geneticamente: 6 foram no gene GCK, 1 no HNF1A, 1 no HNF4A, 1 no HNF1B, 6 em genes raros associados a MODY (1 no ABCC8, 1 no KCNJ11, 1 no PDX1, 2 no PAX4, 1 no NEUROD1), e 1 no NEUROG3, gene associado a diabetes neonatal. Dentre estas 16 variantes, 2 não haviam sido descritas previamente. As 6 mutações no GCK não tinham sido detectadas na análise prévia por: a) 4 casos falso negativos no sequenciamento por Sanger (3 devido ao fenômeno genético de allelic dropout e 1 por erro na leitura do eletroferograma); b) 2 erros na hipótese clínica inicial do subtipo de MODY (baseada no padrão glicêmico e na resposta terapêutica dos pacientes), levando ao sequenciamento prévio de outro gene. A variante no HNF1A não foi detectada previamente por erro na leitura do eletroferograma (caso falso negativo no Sanger). Uma variante foi identificada no gene HNF4A, que não tinha sido sequenciado anteriormente e apresenta fenótipo semelhante ao do HNF1A. O paciente com variante no HNF1B não apresentava relato prévio de cistos renais ou malformações genito-urinárias e por isso não tinha sido considerada a hipótese clínica de MODY5. Além disso, o SLE confirmou o diagnóstico genético de 6 pacientes com variantes em genes de MODY considerados raros, que habitualmente não são sequenciados na rotina de Sanger e ainda detectou uma variante em um gene de diabetes neonatal (sendo necessário maiores estudos para estabelecer uma relação causal com MODY). Em 13 dos 16 casos índices diagnosticados, os familiares encontravam-se disponíveis para exame genético e a co-segregação foi concordante em 8 famílias. Todos os probandos avaliados apresentavam características clínico-laboratoriais típicas de MODY. Os achados deste estudo mostraram que o SLE foi capaz de aumentar a acurácia no diagnóstico de MODY, permitindo a confirmação molecular de 20% dos casos antes negativos e reduzindo, assim, o número de casos MODY X no Brasil. A abordagem genética por painel de SLE para diagnosticar casos com suspeita clínica de MODY mostrou-se promissora para elucidar as bases genéticas desse tipo de diabetes monogênico / Diabetes mellitus type MODY (maturity-onset diabetes of the young) is characterized by defects in insulin secretion, autosomal dominant inheritance, early onset of hyperglycemia, and negative anti-beta cell antibodies. To date, mutations in 14 genes are associated with MODY. The definitive diagnosis relies on genetic tests, traditionally by Sanger sequencing. However, given the genetic heterogeneity of this condition, added to the difficulty of studying some genes due to their large size and lack of hotspots, large-scale sequencing (LSS) seems promising for cost-effective genetic analysis on suspicion of MODY. In Brazil, there are few cohorts screened for MODY and a high prevalence of MODY X (unclear genetic diagnosis). This study aimed to analyze simultaneously all MODY genes in a cohort of clinically suspected patients using a LSS panel; and to evaluate the pathogenicity of identified allelic variants according to the criteria of the American College of Medical Genetics and Genomics (ACMG). We selected 80 subjects with MODY phenotype and negative previous analysis of the 2 most prevalent genes, GCK and HNF1A, by Sanger sequencing method. These cases were analyzed by LSS method, with simultaneous sequencing of target genes. We designed a customized panel, including 51 nuclear genes and the mitochondrial genome. The identified mutations were correlated to the phenotype and family segregation was evaluated. At least 20x coverage was obtained in 98% of the targeted regions. Of 80 evaluated subjects, pathogenic/probably pathogenic variants were detected in 16 cases (20%), confirming the genetic diagnosis of MODY. In 15 of 80 patients, 16 variants of uncertain significance were identified, remaining 42 cases with unexplained molecular diagnosis. Of the 16 genetically confirmed cases: 6 were in the GCK gene, 1 in HNF1A, 1 in HNF4A, 1 in HNF1B, and 6 in rare genes associated with MODY (1 in ABCC8, 1 in KCNJ11, 1 in PDX1, 2 in PAX4 and 1 in NEUROD1), and 1 in NEUROG3, a gene associated with neonatal diabetes. Of these 16 variants, 2 had not been previously described. Those 6 variants in GCK were not detected in the prior analysis because of: a) 4 false negative cases in Sanger sequencing (allelic dropout had occurred in 3 cases and one variant was overlooked, due to electropherogram interpretation failure); b) 2 errors in the initial clinical hypothesis of the MODY subtype (based on the glycemic pattern and therapeutic response), leading to the prior sequencing of another gene. The variant in HNF1A was not previously identified due to misinterpretation in electropherogram (Sanger false negative case). One variant were detected in the HNF4A gene, not formerly sequenced, and had a similar phenotype to that of HNF1A. The patient with HNF1B variant did not have a previous report of renal cysts or genito-urinary malformations and therefore the clinical hypothesis of MODY5 was not considered. In addition, LSS confirmed the genetic diagnosis of 6 patients harboring variants in MODY genes considered to be rare, which are not usually sequenced in the Sanger routine, and also detected one variant in a neonatal diabetes gene (further studies are necessary to establish a causal relationship with MODY). Relatives were available for genetic testing in 13 of these 16 index cases diagnosed and co-segregation was concordant in 8 families. All probands evaluated showed typical clinical and laboratory characteristics of MODY. These study findings showed that targeted-LSS could increase accuracy in MODY diagnosis, enabling molecular confirmation of 20% of previous negative cases and thus reducing the number of MODY X cases in Brazil. The genetic approach of LSS panel to diagnose cases with clinical suspicion of MODY has shown promise for elucidating the genetic basis of this type of monogenic diabetes

L'objectif scientifique de ce travail de thèse est d'améliorer l'estimation des flux d'eau à la surface des continents, à l'échelle saisonnière et interannuelle (de quelques années à décennale). En particulier, il s'agit d'étudier l'apport de données satellites, notamment de la future mission SWOT (Surface Water and Ocean Topography, lancement prévu en 2021), pour l'étude de la partie continentale du cycle de l'eau à l'échelle globale, à l'aide du modèle global de surfaces continentales ISBA-TRIP (Intéractions Sol-Biosphère-Atmosphère/Total Runoff and Integrating Pathways). Dans ce travail de thèse, j'explore le potentiel des données d'altimétrie satellitaire, pour corriger certains paramètres du modèle de routage de rivière TRIP et aussi pour corriger ses variables d'état. Pour ce faire, une plateforme d'assimilation de données virtuelles SWOT, mais aussi de données d'altimètres nadirs actuels a été mise en place. Mais avant l'assimilation de ces données de télédétection, il a été nécessaire de faire une analyse de sensibilité du modèle TRIP à ses paramètres, pour déterminer quels paramètres ont le plus d'influence sur les observables SWOT et qui donc pourront être corrigés. L'analyse de sensibilité (ANOVA) a alors été menée sur les principaux paramètres de TRIP. L'analyse de sensibilité a été menée sur le bassin de L'Amazone et les résultats ont été publiés. Les résultats ont montré que les hauteurs d'eau simulées sont sensibles aux paramètres géomorphologiques locaux exclusivement tandis que les débits simulés sont sensibles à l'ensemble des paramètres amont (selon le réseau de routage TRIP) et surtout au paramètre lié au temps de résidence des eaux souterraines. Enfin, les anomalies de hauteurs présentent des sensibilités similaires aux hauteurs d'eau mais avec des variations temporelles plus marquées. Ces résultats nous ont permis de faire les choix algorithmiques dans le cadre de l'assimilation de données. Ensuite, je me suis concentrée sur le développement de la maquette d'assimilation de données consistant en un Filtre de Kalman d'Ensemble (EnKF) et permet de faire soit de l'estimation de paramètres, soit de l'estimation d'état. La maquette en " estimation de paramètres " est testée et validée par une série d'expériences jumelles. On a assimilé des pseudo-observations de hauteurs et d'anomalies d'eau le long des traces du satellite SWOT, afin de corriger les coefficients de Manning du lit de la rivière, avec possibilité d'étendre à d'autres paramètres. Les premiers résultats montrent que la maquette est capable de retrouver la bonne distribution des coefficients de Manning en assimilant les hauteurs d'eau et les anomalies. Pour l'estimation d'état, on réalise des étapes d'assimilation journalières pour corriger le stock d'eau initial (condition initiale du modèle), en assimilant des débits estimés à partir de séries altimétriques de côtes d'eau ENVISAT. A partir de courbe de tarage hauteurs d'eau-débits calibrées sur le bassin de l'Amazone avec le modèle hydrologique MGB-IPH, les côtes d'eau ont été transformées en " débits altimétriques " que l'on assimile alors dans la maquette. Ces expériences d'estimation d'état nous permettent de sortir du cadre idéalisé des expériences jumelles en assimilant des données réelles, mais nous permet aussi de tester l'apport d'un premier jeu de données de débits provenant de mesures satellites, qui préfigure le futur produit de débit SWOT. Les résultats montrent que les erreurs sur le débits sont globalement améliorées : le run libre donne un RMSE de 2,79x103 m3/s (73,6 %) par rapport aux données in situ disponible sur le bassin et le run corrigé un RMSE de 1,98 x 103 m3/s (53,9 %). / Scientific objective of this PhD work is to improve water fluxes estimation on the continental surfaces, at interanual and interseasonal scale (from few years to decennial time period). More specifically, it studies contribution of remotely-sensed measurements to improve hydrology model. Notably, this work focuses on the incoming SWOT mission (Surface Water and Ocean Topography, launch scheduled for 2021) for the study of the continental water cycle at global scale, and using the land surface model ISBA-TRIP. In this PhD work, I explore the potential of satellite data to correct both input parameters of the river routing scheme TRIP and its state variables. To do so, a data assimilation platform has been set to assimilate SWOT virtual observation as well as discharge estimated from real nadir altimetry data. Beforehand, it was necessary to do a sensibility analysis of TRIP model to its parameters. The aim of such study was to highlight what are the most impacting parameters on SWOT-observed variables and therefore select the ones to correct via data assimilation. The sensibility analysis (ANOVA) has been led on TRIP main parameters. The study has been done over the Amazon basin. The results showed that the simulated water levels are sensitive to local geomorphological parmaters exclusively. On the other hand, the simulated discharges are sensitive to upstream parameters (according to the TRIP river routing network) and more particularly to the groundwater time constant. Finally, water anomalies present sensitivities similar to those of the water levels but with more pronounced temporal variations. These results also lead me to do some choices in the implementation of the assimilation scheme and have been published. Therefore, in the second part of my PhD, I focused on developing a data assimilation platform which consists in an Ensemble Kalman Filter (EnKF). It could either correct the model input parameters or directly its state. A series of twin experiments is used to test and validate the parameter estimation module of the platform. SWOT virtual-observations of water heights and anomalies along SWOT tracks are assimilated to correct the river manning coefficient, with the possibility to easily extend to other parameters. First results show that the platform is able to recover the "true" Manning distribution assimilating SWOT-like water heights and anomalies. In the state estimation mode, daily assimilation cycles are realized to correct TRIP river water storage initial state by assimilating ENVISAT-based discharge. Those observations are derived from ENVISAT water elevation measures, using rating curves from the MGB-IPH hydrological model (calibrated over the Amazon using in situ gages discharge). Using such kind of observation allows going beyond idealized twin experiments and also to test contribution of a remotely-sensed discharge product, which could prefigure the SWOT discharge product. The results show that discharge after assimilation are globally improved : the root-mean-square error between the analysis discharge ensemble mean and in situ discharges is reduced by 28 \%, compared to the root-mean-square error between the free run and in situ discharges (RMSE are respectively equal to 2.79 x 103 m3/s and 1.98 x 103 m3/s).