Proteomics serves an important role at the systems-level in understanding of biological functioning. Mass spectrometry proteomics has become the tool of choice for identifying and quantifying the proteome of an organism. In the most widely used bottom-up approach to MS-based high-throughput quantitative proteomics, complex mixtures of proteins are first subjected to enzymatic cleavage, the resulting peptide products are separated based on chemical or physical properties and then analyzed using a mass spectrometer. The three fundamental challenges in the analysis of bottom-up MS-based proteomics are as follows: (i) Identifying the proteins that are present in a sample, (ii) Aligning different samples on elution (retention) time, mass, peak area (intensity) and etc, (iii) Quantifying the abundance levels of the identified proteins after alignment. Each of these challenges requires knowledge of the biological and technological context that give rise to the observed data, as well as the application of sound statistical principles for estimation and inference. In this dissertation, we present a set of statistical methods in bottom-up proteomics towards protein identification, alignment and quantification. We describe a fully Bayesian hierarchical modeling approach to peptide and protein identification on the basis of MS/MS fragmentation patterns in a unified framework. Our major contribution is to allow for dependence among the list of top candidate PSMs, which we accomplish with a Bayesian multiple component mixture model incorporating decoy search results and joint estimation of the accuracy of a list of peptide identifications for each MS/MS fragmentation spectrum. We also propose an objective criteria for the evaluation of the False Discovery Rate (FDR) associated with a list of identifications at both peptide level, which results in more accurate FDR estimates than existing methods like PeptideProphet. Several alignment algorithms have been developed using different warping functions. However, all the existing alignment approaches suffer from a useful metric for scoring an alignment between two data sets and hence lack a quantitative score for how good an alignment is. Our alignment approach uses "Anchor points" found to align all the individual scan in the target sample and provides a framework to quantify the alignment, that is, assigning a p-value to a set of aligned LC-MS runs to assess the correctness of alignment. After alignment using our algorithm, the p-values from Wilcoxon signed-rank test on elution (retention) time, M/Z, peak area successfully turn into non-significant values. Quantitative mass spectrometry-based proteomics involves statistical inference on protein abundance, based on the intensities of each protein's associated spectral peaks. However, typical mass spectrometry-based proteomics data sets have substantial proportions of missing observations, due at least in part to censoring of low intensities. This complicates intensity-based differential expression analysis. We outline a statistical method for protein differential expression, based on a simple Binomial likelihood. By modeling peak intensities as binary, in terms of "presence / absence", we enable the selection of proteins not typically amendable to quantitative analysis; e.g., "one-state" proteins that are present in one condition but absent in another. In addition, we present an analysis protocol that combines quantitative and presence / absence analysis of a given data set in a principled way, resulting in a single list of selected proteins with a single associated FDR.

O final da vida útil de pavimentos com bases rígidas se caracteriza pelo aparecimento de trincas de blocos e de fadiga. Neste cenário, a reciclagem das camadas de base e revestimento asfáltico, com adição de cimento, surge como técnica importante para reabilitação dessas estruturas. Além de se tratar de uma solução vantajosa do ponto de vista técnico, é competitiva em termos econômicos, além de sustentável. Entretanto, a escassa normatização nacional dificulta sua aplicação, destacando-se a falta de um procedimento de dosagem. Objetivando contribuir para o desenvolvimento de um método de dosagem de camadas recicladas com cimento, desenvolveu-se um programa experimental, contemplando a caracterização mecânica (resistência e rigidez), a variação volumétrica e a erodibilidade de misturas contendo fresado asfáltico, materiais de bases rígidas (brita graduada tratada com cimento e solo-cimento) e cimento Portland. Foram analisados os efeitos da porcentagem de fresado, do teor de cimento e do tempo de cura, compactando-se os corpos de prova na energia Modificada. Com auxílio de um software estatístico, foi elaborado um planejamento experimental para definição das misturas analisadas, contemplando-se diversos níveis para as variáveis independentes (teores de cimento entre 1% e 7%, e porcentagens de fresado variando entre 8% e 92%, aproximadamente). Os tempos de cura considerados foram extrínsecos ao planejamento e variaram de acordo com o ensaio, sendo de 3, 7 e 14 dias para os ensaios de comportamento mecânico e de 7 dias para os demais ensaios. Obtiveram-se modelos com efeitos estatisticamente significativos, com exceção da expansão. Os demais modelos apresentaram coeficientes de determinação de médios a elevados, tendo em vista a heterogeneidade dos materiais estudados, além do tamanho amostral considerável. Foi verificado que todas as variáveis afetam as propriedades analisadas, sendo que o teor de cimento demonstrou maior efeito, sendo que sua adição melhora consideravelmente o comportamento das misturas empregadas, frente à ação do tráfego e da água. Foram alcançados resultados bastante elevados de resistência à compressão simples (1,00 MPa a 6,49 MPa) e resistência à tração na compressão diametral (0,17 MPa a 1,22 MPa); já a rigidez das misturas apresentou uma ampla variação de resultados (484 MPa a 20.031 MPa). Com relação aos materiais de base empregados (brita graduada tratada com cimento e solo-cimento), não se observou uma tendência única; as misturas com solo-cimento apresentaram maior resistência, entretanto, com comportamento ligeiramente inferior quanto à variação volumétrica e à erodibilidade. Na análise da rigidez, os materiais de base mostraram comportamento bastante similar. Para concluir, foi verificado que o procedimento proposto por Fedrigo (2015) também é satisfatório para a dosagem de misturas recicladas constituídas por antigas bases cimentadas e fresado asfáltico, restando ainda quantificar a possível retração por secagem das misturas estudadas. / The end of the useful life of pavements with rigid bases characterizes by the appearance of block and fatigue cracking. The full-depth reclamation with cement (FDR-C) of this layer combined with the asphalt layer seem to be a good choice for the rehabilitation of the pavement structure. Besides being an advantageous solution from a technical point of view, it is competitive in economic terms, besides being sustainable. However, the Brazilian standards are scarce thus limiting its application, highlighting the lack of an appropriated mix design method to this technique. In order to contribute for the development of an FDR-C mix design method, an experimental program was developed aiming to test the mechanical characterization (strength and stiffness), volumetric variation and the erodibility of mixtures made of reclaimed asphalt pavement (RAP), rigid base materials (cement treated crushed stone and soil-cement) and Portland cement. There were analyzed the effects of the RAP percentage, the cement content and the curing time of the specimens. For that, the compaction effort used was the Brazilian Modified one. Through a statistic software, an experimental planning was prepared for mixtures determination, with several levels for the independent variables (cement content between 1% and 7%, and RAP percentage varying by 8% up to 92%, approximately). The curing time analyzed was outward of experimental planning and it varied according to the kind of test. The specimens of mechanical performance were cured for 3, 7 and 14 days, and the other tests were examined at the 7th day of curing time. Models with statically significant effects were obtained, except the swell one. The others had medium and high coefficients of determination, given the heterogeneity of the studied materials, in addition to considerable sample size. It was found that all variables affect the analyzed properties, and the cement content proved to be with the greatest effect among the analyzed factors. Cement addition improves considerably the behavior of the mixtures for the traffic and water actions. Moreover, higher results were achieved for UCS (1.00 MPa up to 6.49 MPa) and ITS (0.17 MPa up to 1.22 MPa) tests. While the mixtures stiffness presented a wide range of results (484 MPa up to 20,031 MPa). Regarding the studied materials, it was observed that they do not have a single trend. The mixtures with soil-cement presented stronger, but with slightly lower behavior for volumetric variation and erodibility. At the stiffness analysis, different materials showed very similar behavior. All in all, it was found that the mix design method proposed by Fedrigo (2015) is also suitable for employment in semi-rigid FDR-C, still remaining to measure the possible drying shrinkage of specimens.

O final da vida útil de pavimentos com bases rígidas se caracteriza pelo aparecimento de trincas de blocos e de fadiga. Neste cenário, a reciclagem das camadas de base e revestimento asfáltico, com adição de cimento, surge como técnica importante para reabilitação dessas estruturas. Além de se tratar de uma solução vantajosa do ponto de vista técnico, é competitiva em termos econômicos, além de sustentável. Entretanto, a escassa normatização nacional dificulta sua aplicação, destacando-se a falta de um procedimento de dosagem. Objetivando contribuir para o desenvolvimento de um método de dosagem de camadas recicladas com cimento, desenvolveu-se um programa experimental, contemplando a caracterização mecânica (resistência e rigidez), a variação volumétrica e a erodibilidade de misturas contendo fresado asfáltico, materiais de bases rígidas (brita graduada tratada com cimento e solo-cimento) e cimento Portland. Foram analisados os efeitos da porcentagem de fresado, do teor de cimento e do tempo de cura, compactando-se os corpos de prova na energia Modificada. Com auxílio de um software estatístico, foi elaborado um planejamento experimental para definição das misturas analisadas, contemplando-se diversos níveis para as variáveis independentes (teores de cimento entre 1% e 7%, e porcentagens de fresado variando entre 8% e 92%, aproximadamente). Os tempos de cura considerados foram extrínsecos ao planejamento e variaram de acordo com o ensaio, sendo de 3, 7 e 14 dias para os ensaios de comportamento mecânico e de 7 dias para os demais ensaios. Obtiveram-se modelos com efeitos estatisticamente significativos, com exceção da expansão. Os demais modelos apresentaram coeficientes de determinação de médios a elevados, tendo em vista a heterogeneidade dos materiais estudados, além do tamanho amostral considerável. Foi verificado que todas as variáveis afetam as propriedades analisadas, sendo que o teor de cimento demonstrou maior efeito, sendo que sua adição melhora consideravelmente o comportamento das misturas empregadas, frente à ação do tráfego e da água. Foram alcançados resultados bastante elevados de resistência à compressão simples (1,00 MPa a 6,49 MPa) e resistência à tração na compressão diametral (0,17 MPa a 1,22 MPa); já a rigidez das misturas apresentou uma ampla variação de resultados (484 MPa a 20.031 MPa). Com relação aos materiais de base empregados (brita graduada tratada com cimento e solo-cimento), não se observou uma tendência única; as misturas com solo-cimento apresentaram maior resistência, entretanto, com comportamento ligeiramente inferior quanto à variação volumétrica e à erodibilidade. Na análise da rigidez, os materiais de base mostraram comportamento bastante similar. Para concluir, foi verificado que o procedimento proposto por Fedrigo (2015) também é satisfatório para a dosagem de misturas recicladas constituídas por antigas bases cimentadas e fresado asfáltico, restando ainda quantificar a possível retração por secagem das misturas estudadas. / The end of the useful life of pavements with rigid bases characterizes by the appearance of block and fatigue cracking. The full-depth reclamation with cement (FDR-C) of this layer combined with the asphalt layer seem to be a good choice for the rehabilitation of the pavement structure. Besides being an advantageous solution from a technical point of view, it is competitive in economic terms, besides being sustainable. However, the Brazilian standards are scarce thus limiting its application, highlighting the lack of an appropriated mix design method to this technique. In order to contribute for the development of an FDR-C mix design method, an experimental program was developed aiming to test the mechanical characterization (strength and stiffness), volumetric variation and the erodibility of mixtures made of reclaimed asphalt pavement (RAP), rigid base materials (cement treated crushed stone and soil-cement) and Portland cement. There were analyzed the effects of the RAP percentage, the cement content and the curing time of the specimens. For that, the compaction effort used was the Brazilian Modified one. Through a statistic software, an experimental planning was prepared for mixtures determination, with several levels for the independent variables (cement content between 1% and 7%, and RAP percentage varying by 8% up to 92%, approximately). The curing time analyzed was outward of experimental planning and it varied according to the kind of test. The specimens of mechanical performance were cured for 3, 7 and 14 days, and the other tests were examined at the 7th day of curing time. Models with statically significant effects were obtained, except the swell one. The others had medium and high coefficients of determination, given the heterogeneity of the studied materials, in addition to considerable sample size. It was found that all variables affect the analyzed properties, and the cement content proved to be with the greatest effect among the analyzed factors. Cement addition improves considerably the behavior of the mixtures for the traffic and water actions. Moreover, higher results were achieved for UCS (1.00 MPa up to 6.49 MPa) and ITS (0.17 MPa up to 1.22 MPa) tests. While the mixtures stiffness presented a wide range of results (484 MPa up to 20,031 MPa). Regarding the studied materials, it was observed that they do not have a single trend. The mixtures with soil-cement presented stronger, but with slightly lower behavior for volumetric variation and erodibility. At the stiffness analysis, different materials showed very similar behavior. All in all, it was found that the mix design method proposed by Fedrigo (2015) is also suitable for employment in semi-rigid FDR-C, still remaining to measure the possible drying shrinkage of specimens.

Made available in DSpace on 2014-08-20T14:33:08Z (GMT). No. of bitstreams: 1 Dissertacao_Viviane_Santos_Silva_Terra.pdf: 2073573 bytes, checksum: 19ddda03c961d08b0c0179b761552932 (MD5) Previous issue date: 2010-02-12 / The South region of Brazil (Pelotas, state of Rio Grande do Sul) is responsible for the production of temperate fruit trees. Among them, the peach crop has been a good option for crops diversification. The knowledge of the amount of available soil water to supply water needs during the peach different stages of development is of interest to the agronomic point of view. In this sense, we highlight the importance of the use of irrigation in peach orchards in order to provide an adequate crop water supply. It is therefore important to know the response of peach to water deficit so that we can assess the influence of the practice of irrigation in crop production. Among the methodologies, the soil water balance, which consists of applying the equation of conservation of mass, has been used. Based on this, the main objective of this work was to evaluate and quantify the different components of water balance in a Red Yellow Podzolic soil, planted with peach trees, cultivar Maciel, under irrigation and without irrigation and to verify the adequacy of Domain Reflectometry Frequency (FDR) equipment for monitoring the water content over time. The evaluation and quantification of different components of water balance of soil water were measured in an experimental area which belongs to Embrapa Clima Temperado, Pelotas-RS, Brazil, planted with peach cultivar Maciel, nearby plants irrigated and not irrigated. For this, the water balance calculation was carried out weekly and divided into 32 periods, of which 4 periods (12, 20, 24 and 26) were discarded because they had precipitation above 30 mm week-1. The experimental area containing irrigated and no irrigated plants was divided into five strips. The irrigated plants belonged to the strip 1 (tubes 3 and 6), strip 2 (tube 2) and strip 3 (tube 1), each one with an area of 0.5 m². The no-irrigated area belonged to the strip 4 (tube 8) and 5 (tube 7), with an area of 1 meters each one. Strips 1, 2 and 4 were under canopy and the other outside of the canopy. Rainfall was measured by rain gauges; irrigation water supply was made based on the average daily evapotranspiration of the region; the variation of water storage in the soil was measured by readings of soil water content up to 0.50 m depth, using an equipment based on reflectometry in the frequency domain (FDR) which was calibrated from readings of tensiometers; the runoff and internal drainage or capillary rise were neglected; and the actual evapotranspiration (ETr) was taken as the unknown variable. From the results we can conclude that it is possible to use an indirect method, using tensiometers and soil water retention curves to obtain calibration curves of the reflectometry equipment in the frequency domain (FDR); the adopted irrigation management, based on the average evapotranspiration of the region, may not be sufficient to supply the water requirements of peach crop; peach trees with water supply through irrigation have higher values of actual evapotranspiration; and soil water storage ranged in all irrigated and no irrigated strips between 60 and 140 mm week-1, unaffected by the adopted treatments in the study. / A região de Pelotas é responsável pela grande produção de frutíferas de Clima Temperado, em especial a cultura do pessegueiro que tem sido uma boa opção para a diversificação de culturas. O conhecimento da quantidade de água disponível no solo para atender as necessidades hídricas do pessegueiro durante os seus diferentes estádios de desenvolvimento é de interesse do ponto de vista agronômico. Neste sentido, destaca-se o uso da irrigação na cultura do pessegueiro, visando fornecer a cultura um adequado suprimento de água. Desta forma, é importante conhecer a resposta do pessegueiro ao déficit hídrico para que se possa avaliar a influência da prática da irrigação na produção da cultura. Dentre as metodologias que tem sido utilizada, destaca-se a do balanço hídrico no solo, que consiste na aplicação da equação de conservação de massa. Baseado no acima exposto, este trabalho teve como o objetivo principal avaliar e quantificar os diferentes componentes do balanço hídrico em um solo Argissolo Vermelho Amarelo, cultivado com plantas de pessegueiro, cultivar Maciel, em condições de irrigação e sem irrigação, bem como verificar a adequação de um equipamento de Reflectometria no Domínio da Frequência (FDR) para monitorar o seu conteúdo de água ao longo do tempo. A avaliação e quantificação dos diferentes componentes do balanço hídrico de água no solo foram realizadas em uma área experimental pertencente a Embrapa Clima Temperado, Pelotas-RS, cultivada com pessegueiro, cultivar Maciel, próximo às plantas irrigadas e não irrigadas. Para tal, o cálculo do balanço hídrico foi realizado semanalmente e dividido em 32 períodos, sendo a área experimental contendo as plantas irrigadas e não irrigadas dividida em cinco faixas. As plantas irrigadas pertenciam a faixa 1 (tubos 3 e 6), faixa 2 (tubo 2) e faixa 3 (tubo 1), cada uma com área de 0,5 m². As não irrigadas pertenciam a faixa 4 (tubo 8) e faixa 5 (tubo 7), com uma área de 1,0 m² cada. As faixas 1, 2 e 4 encontravamse na subcopa da planta e as demais fora da copa. A precipitação foi medida por meio de pluviômetros; a irrigação foi feita diariamente baseada na média evapotranspirativa da região; a variação da armazenagem de água no solo foi medida por meio de leituras de umidade do solo até 0,50 m de profundidade, utilizando-se um equipamento baseado na reflectometria no domínio da frequência (FDR), o qual foi calibrado a partir de leituras de tensiometria; o escoamento superficial e a drenagem interna ou ascensão capilar foram desprezados; a evapotranspiração foi tomada como variável incógnita. A partir dos resultados podese concluir que é possível a utilização de um método indireto, usando tensiômetros e curvas de retenção para a obtenção de curvas de calibração do equipamento de reflectometria no domínio da frequência (FDR); o manejo de irrigação adotado, baseado na média de evapotranspiração da região, pode não ser suficiente para suprir a demanda hídrica da cultura do pessegueiro; plantas de pessegueiro com suprimento hídrico por meio da irrigação apresentam valores maiores de evapotranspiração real e a armazenagem variou em todas as faixas irrigadas e não irrigadas entre 60 a 140 mm semana-1, não sendo afetada pelos manejos adotados no estudo.

The aim of the master's thesis "Analysis of Selected Payment Protocols" is overview of used payment. The first part is concentrated on data security, the second is dedicated to payment protocols, their characteristics, used technology and security elements. The third part is dedicated to verification and simulation tools. Comparison of particular payment protocols and of particular verification tools is part of this work too. Experimental part of the thesis is focused on formalization and verification of the payment protocol Visa 3-D Secure, of the protocol NetBill and on formalization of two subprotocols of SET.

The purpose of this research was to determine the effect of full-depth reclamation (FDR) on the strength and durability of aggregate base layers in a coordinated approach involving both field and laboratory testing. Field comparisons between the pre-reclamation neat base and post-reclamation blended base were supplemented with laboratory experiments conducted to determine the effects of reclaimed asphalt pavement (RAP) content, compaction effort, and heating on the strength and durability of roadways reconstructed using FDR with a portable asphalt recycling machine (PARM). Also, the effect of reclamation on the spatial uniformity of the pavement structures was explored by comparing variability in the pre- and post-reclamation material properties. Test sites in Orem, Utah; San Marcos, Texas; and South Jordan, Utah, were selected for this research. The results of field testing indicate that the FDR process significantly increased the stiffness and/or strength of the base material at two of the test locations and did not significantly change the third base material. An evaluation of spatial variability indicated that the FDR process produced equivalent or lower spatial variability with respect to both base modulus and California bearing ratio (CBR) values at one site, while the other two sites exhibited equivalent or higher spatial variability after FDR. The results of laboratory testing for all three locations indicate that specimens compacted using the modified Proctor method exhibit significantly higher CBR values and dry densities than specimens compacted using the standard Proctor method. Also, the CBR values for specimens tested in the dry condition were significantly higher than those obtained from specimens tested at optimum moisture content. These results demonstrate the value of achieving a high level of compaction during construction and preventing water ingress into the pavement over time. The blended material exhibited a significantly lower CBR value than that of the neat material at only one location; the addition of RAP to materials at the other locations did not significantly change the CBR values of those materials. In the tube suction test (TST), most of the specimens were classified as marginally or highly moisture-susceptible, and the effect of RAP on the dielectric value in the TST was of no practical importance. The use of PARMs in the FDR process is an acceptable, economical, and environmentally friendly approach to reconstruction of flexible pavements. To ensure satisfactory performance of FDR projects, engineers and managers should carefully follow recommended guidelines for project selection, pavement testing, material characterization, design, construction, and quality assurance testing.

The FDR (Frequency Disturbance Recorder) is a data acquisition device for the power system. The device is portable and can be used with any residential wall outlet for frequency data collection. Furthermore, the FDR transmits calculated frequency data to the web for access by authorized users via Ethernet connection. As a result, Virginia Tech implemented Frequency Monitoring Network (FNET) with these FDR devices. FNET is a collection of identical FDRs placed in different measurement sites to allow for data integration and comparison. Frequency is an important factor for power system control and stabilization. With funding and support provided by ABB, TVA and NSF the FDRs are placed strategically all over the United States for frequency analysis, power system protection and monitoring. The purpose of this study is to refine the current FDR hardware design and establish a new design that will physically fit all the components on one Printed Circuit Board (PCB). At the same time, the software that is to be implemented on the new board is to be kept similar if not the same as that of the current design. The current FDR uses the Axiom CME555 development board and it is interfaced to the external devices through its communication ports. Even through the CME555 board is able to meet the demands of the basic FDR operations, there are still several problems associated with this design. This paper will address some of those hardware problems, as well as propose a new board design that is specifically aimed for operations of FDR. / Master of Science

在DNA微陣列基因的實驗中資料包括數千個cDNA 序列，為了要篩選出有差異表現基因，同時針對大量基因個數作假設檢定。若無適當地調整個別檢定問題中的誤差率，則將會膨脹整體的誤差率。在多重假設檢定中為了讓整體誤差率(familywise error rate, FWE)控制在設定水準下，必須調整個別假設檢定之個別型一誤差率CWE的檢定準則，此為多重比較方法(multiple comparison procedures：MCP)。然而當多重比較的個數增加時，控制整體誤差率FWE之傳統的多重比較方法會是過於嚴格的標準，不容易推翻虛無假設，使得檢定的結果太過保守。為了解決此現象，Benjamini and Hochberg(1995) 建議另一種錯誤率：錯誤發現率(false discovery rate：FDR)。錯誤發現率定義為在被拒絕之虛無假設中錯誤拒絕的比例之期望值。而Benjamini and Hochberg(1995)也在文中提出一個得以控制錯誤發現率的多重比較方法，稱為BH方法。本篇論文將詳盡地介紹CWE、FWE和FDR三種誤差率，並提出－修正BH的方法，稱為BH( )。我們將透過電腦模擬驗證出新的修正BH方法之表現比原BH方法有較高的檢定力，且從實例的結果中發現BH( )比原BH方法能檢測出更多的顯著個數。 關鍵字：個別型一誤差率(CWE)；整體誤差率(FWE)；多重比較方法(MCP)； 錯誤發現率(FDR)。 / cDNA microarray technology provides tools to study thousands of genes simultaneously. Since a large number of genes are compared, using a conventional significant test leads to the increase of the type I error rate. To avoid the inflation, the adjustment for multiplicity should be considered and a multiple comparison procedure (MCP) that controls the familywise error rate (FWE) is recommended. However, the conservativeness of a MCP that controls FWE becomes more and more severe as the number of comparisons (genes) increases. Instead of FWE, Benjamini and Hochberg (1995) recommended to control the expected proportion of falsely rejecting hypotheses—the false discovery rate (FDR)—and developed a MCP, which has its FDR under control. In this paper, the error rates CWE, FWE and FDR are fully introduced. A new MCP with FDR controlled is developed and its performance is investigated through intensive simulations. KEY WORDS：Comparison-wise error rate (CWE)；Familywise error rate (FWE)；Multiple comparison procedure (MCP)；False discovery rate (FDR).

Neste trabalho apresenta-se uma revisão da metodologia de experimentos de microarray relativas a sua instalação e análise estatística dos dados obtidos. A seguir, aplica-se essa metodologia na análise de dados de expressão gênica em citros, gerados por um experimento de macroarray, utilizando modelos lineares de efeitos fixos considerando a inclusão ou não de diferentes efeitos e considerando ajustes de modelos para cada gene separadamente e para todos os genes simultaneamente. Os experimentos de macroarray são similares aos experimentos de microarray, porém utilizam um menor número de genes. Em geral, são utilizados devido a restrições econômicas. Devido ao fato de terem sido utilizados poucos arrays no experimento analisado neste trabalho foi utilizada uma abordagem bayesiana empírica que utiliza estimativas de variância mais estáveis e que leva em consideração a correlação entre as repetições do gene dentro do array. Também foi utilizado um método de análise não paramétrico para contornar o problema da falta de normalidade para alguns genes. Os resultados obtidos em cada um dos métodos de análise descritos foram então comparados. / This paper presents a review of the methodology of microarray experiments for its installation and statistical analysis of data obtained. Then this methodology is applied in data analysis of gene expression in citrus, generated by a macroarray experiment, using linear models with fixed effects considering the inclusion or exclusion of different effects and considering adjustments of models for each gene separately and for all genes simultaneously. The macroarray experiments are similar to the microarray experiments, but use a smaller number of genes. In general, are used due to economic restrictions. Because they have been used a few arrays in the experiment analyzed in this study it was used a empirical Bayes approach that uses estimates of variance more stable and that takes into account the correlation among replicates of the gene within array. A non parametric analysis method was also used to outline the problem of the non normality for some genes. The results obtained in each of the described methods of analysis were then compared.

L'objet de cette thèse est l'étude d'un certain type d'algorithmes de rééchantillonnage regroupés sous le nom de validation-croisée, et plus particulièrement parmi eux, du leave-p-out. Très utilisés en pratique, ces algorithmes sont encore mal compris d'un point de vue théorique, notamment sur un plan non-asymptotique. Notre analyse du leave-p-out s'effectue dans les cadres de l'estimation de densité et de la régression. Son objectif est de mieux comprendre la validation-croisée en fonction du cardinal $p$ de l'ensemble test dont elle dépend. D'un point de vue général, la validation-croisée est destinée à estimer le risque d'un estimateur. Dans notre cas, le leave-$p$-out n'est habituellement pas applicable en pratique, à cause d'une trop grande complexité algorithmique. Pourtant, nous parvenons à obtenir des formules closes (parfaitement calculables) de l'estimateur leave-p-out du risque, pour une large gamme d'estimateurs très employés. Nous envisageons le problème de la sélection de modèle par validation-croisée sous deux aspects. L'un repose sur l'estimation optimale du risque en termes d'un compromis biais-variance, ce qui donne lieu à une procédure d'estimation de densité basée sur un choix de $p$ entièrement fondé sur les données. Une application naturelle au problème des tests multiples est envisagée. L'autre aspect est lié à l'interprétation de l'estimateur validation-croisée comme critère pénalisé. Sur le plan théorique, la qualité de la procédure leave-$p$-out est garantie par des inégalités oracle ainsi qu'un résultat d'adaptativité dans le cadre de l'estimation de densité. Le problème de la détection de ruptures est également abordé au travers d'une vaste étude de simulations, basée sur des considérations théoriques. Sur cette base, nous proposons une procédure entièrement tournée vers le rééchantillonnage, permettant de traiter le cas difficile de données hétéroscédastiques avec une complexité algorithmique raisonnable.