Conselho Nacional de Desenvolvimento Científico e Tecnológico / The sulfur compounds have been studied for many years due to their importance in chemical, biological and industrial areas. These compounds can also be found in matrices of oil and its derivatives, and among the groups, the most important are the sulfides, thiophenes and benzothiophenes. The determination of these compounds in asphalt has great importance since they are considered precursors of oxidation reactions and are directly related to its aging, which reflects the durability of pavements. Therefore, various methods for determining these compounds have been described, although there is a lack of information in the literature about studies of sulfur compounds employing voltammetry and gold electrode. Thus, the aim of this study was to develop an electrochemical method for determining organo-sulfur compounds in asphalt oil cement samples (CAP) using differential pulse voltammetry in non-aqueous medium, a gold electrode as working electrode, an Ag/AgCl reference electrode modified with LiCl saturated in ethanol and 0.1M NaClO4 in DMSO and a platinum wire as counter electrode, as a supporting electrolyte was used a 0.1M NaCl4 solution in DMSO. / Os compostos de enxofre têm sido estudados há muitos anos devido à sua importância em áreas químicas, biológicas e industriais. Esses compostos podem ser encontrados também em matrizes de petróleo e seus derivados, e entre as principais classes destacam-se os sulfetos, tiofenos e benzotiofenos. A determinação destes compostos em asfaltos é de grande importância uma vez que são considerados precursores das reações de oxidação e estão diretamente relacionados com o seu envelhecimento, o que reflete na durabilidade dos pavimentos. Em vista disso, diferentes métodos para a determinação desses compostos têm sido descritos, porém a literatura não contém muitos relatos de estudos de compostos sulfurados empregando voltametria e eletrodo de ouro. Assim, o objetivo deste trabalho foi desenvolver um método eletroquímico para determinar compostos organo-sulfurados em amostras de cimento asfáltico de petróleo (CAP) utilizando a voltametria de pulso diferencial em meio não aquoso, um eletrodo de ouro como eletrodo de trabalho, um eletrodo de referência de Ag/AgCl modificado com LiCl saturado em etanol e NaClO4 0,1M em DMSO e um fio de platina como contra-eletrodo, como eletrólito de suporte foi utilizado uma solução de NaClO4 0,1M em DMSO.

In this work the compression behavior of masonry was investigated. After a detailed review of code approaches and different research works, a new formula was suggested to describe the compression strength of masonry, based on the mechanical and geometrical properties of its components, when deformation properties of units are larger than the ones of mortar. Later on, a new model, Extended Drucker-Prager Cap Yielding Function, is suggested to describe the three axial compression stress state of mortar in masonry in case deformation properties of mortar are larger than the ones of mortar, and to describe the three axial compression stress state of brick in the other case. This includes defining its parameters based on test diagrams of the mortar material, implementing the model in the numerical software ANSYS, and the numerical results are evaluated for simple cube example. The controlling equations of creep based on the visco-elastic creep theory are presented in the general case of three axial creep under three axial loading conditions. The special case of three axial creep under axial loading is also presented. The “transversal creep” relevant for the compression strength of masonry was discussed and numerical examples have been added to show the effect of changed time-dependent Poisson’s ratio. In another chapter, many examples are presented showing the application of the suggested material models and discontinuous numerical method named eXtended finite element method. Conclusions and recommendations are given in the last chapter.

In this work the compression behavior of masonry was investigated. After a detailed review of code approaches and different research works, a new formula was suggested to describe the compression strength of masonry, based on the mechanical and geometrical properties of its components, when deformation properties of units are larger than the ones of mortar. Later on, a new model, Extended Drucker-Prager Cap Yielding Function, is suggested to describe the three axial compression stress state of mortar in masonry in case deformation properties of mortar are larger than the ones of mortar, and to describe the three axial compression stress state of brick in the other case. This includes defining its parameters based on test diagrams of the mortar material, implementing the model in the numerical software ANSYS, and the numerical results are evaluated for simple cube example. The controlling equations of creep based on the visco-elastic creep theory are presented in the general case of three axial creep under three axial loading conditions. The special case of three axial creep under axial loading is also presented. The “transversal creep” relevant for the compression strength of masonry was discussed and numerical examples have been added to show the effect of changed time-dependent Poisson’s ratio. In another chapter, many examples are presented showing the application of the suggested material models and discontinuous numerical method named eXtended finite element method. Conclusions and recommendations are given in the last chapter.

La famille des protéines Fragile X Related (FXR) comprend la protéine FMRP ainsi que les homologues FXR1P et FXR2P. En plus d'une forte homologie de séquence, tous les membres de cette famille de protéines possèdent des domaines caractéristiques aux molécules liant l'ARN ainsi que des signaux d'importation et d'exportation nucléaire. FMRP est l’archétype de cette famille de protéine, puisque son absence cause le retard mental avec X Fragile. Par contre, aucune pathologie n’est associée avec la perte d’expression des homologues FXR1P et FXR2P et ce, même si ces deux protéines ont été mises en évidence dans des processus développementaux chez la souris. En effet, cette famille de protéines semble jouer un rôle primordial durant l’embryogenèse, puisque la délétion de FMR1 et FXR2 provoque des troubles cognitifs, alors que FXR1 semble plutôt jouer un rôle dans la myogenèse et la spermatogenèse chez les mammifères. Cette diversification fonctionnelle de FXR1 semble être attribuable à l’expression complexe de ses différentes isoformes. En effet, chez les mammifères, quatre des six isoformes de FXR1P (70, 74, 78 et 80 kDa) sont exprimées dans tous les tissus à l’exception des muscles striés et cardiaques où elles sont remplacées par deux isoformes dites « muscle spécifique » (82 et 84 kDa). Le nombre élevé d’isoformes de FXR1 rend cette protéine difficile à étudier chez les mammifères. Cette expression de FXR1 est hautement conservée chez tous les vertébrés et peut être décelée chez plusieurs organismes tels le poulet, le poisson zèbre ainsi que chez la grenouille Xenopus laevis. Le xénope s’avère être le modèle exemplaire, puisque l’expression de xFxr1 y est beaucoup plus simple et ce, tout en conservant l’expression tissu spécifique de ces isoformes. En effet, seule une isoforme de 84 kDa est exprimée dans tous les tissus à l’exception des muscles striés et cardiaques où il y a expression d’une isoforme de 88 kDa. Étant donné le rôle dans les cellules musculaires striées, il est impératif de comprendre les implications de l’inactivation de ce gène chez les vertébrés. / Fragile X Mental Retardation Protein (FMRP) is part of a mRNA-binding proteins family that includes the Fragile X Related 1 and 2 proteins (FXR1P and FXR2P). These proteins share multiple functional domains typical of mRNA-binding domain (two KH domains and 1 RGG box) as well as a nuclear and a cytoplasmic localization domain. Whereas absence of FMRP is the cause of Fragile X Mental Retardation in human, it is not known whether FXR1P and FXR2P are associated to any pathology and whether these homologous proteins can compensate for the absence of FMRP in the case of the Fragile X syndrome. Knockout mice for FXR proteins are powerful tools that are commonly used in research to shed light on the functions of these proteins and point out their embryonic involvement. However, the Fxr1 knockout mouse didn’t proved to be a good model as the two mentioned above. In mammals, we have shown that FXR1 play a key role in muscle differentiation, since two of the six isoforms are muscle specific and are believed to be essential for the normal development of the cardiac and skeletal muscle. Although essential for embryonic development, it is nearly impossible to study the developmental implication of the differential expression of these tissues specific proteins in mammals due to the large number of FXR1P isoform. Simpler model such as drosophila melanogaster are being used, but this model have only one proteins (dFMRP) which is expressed ubiquitously in this organism and do not represent the tissue specific expression of some of the family member. We choose an intermediate model such as Xenopus laevis, which is an extensively used model for developmental studies, and proceeded with the inactivation of xFxr1. In Xenopus laevis, we found two different xFxr1 proteins isoform; one short isoform (84 KDa) is ubiquitously expressed in every tissues except in muscle, whereas the long isoform (88 KDa) is expressed only in cardiac and skeletal muscle. Specific inactivation of xFxr1 messengers during the early development gave us new insight on the specific functions of these proteins during the embryogenesis and primary myogenesis.

Au nord-ouest de la mer Méditerranée, dans le Golfe du Lion, la marge continentale est incisée par plusieurs canyons. Leur morphologie suggère que les instabilités jouent un rôle important pour la formation de ces vallées sous-marines et pour le transport de sédiments. Le canyon du Cap de Creus, situé à l'extrémité ouest du Golfe du Lion, a été sélectionné afin d'y analyser les processus d'instabilité présentement actifs. Les profils sismiques qui couvrent le secteur ont permis de déterminer la morphostratigraphie du canyon. Plusieurs essais géotechniques ont été effectués sur les sédiments du secteur d'étude afin de déterminer leurs paramètres géomécaniques et leur état de consolidation. A l'aide de ces résultats, un modèle géologique du flanc a été établi et utilisé pour modéliser la stabilité des flancs et la mobilité des masses instables. Ainsi, l'influence de processus naturels sur le développement des ruptures a été déterminée. La propagation des débris a ensuite été évaluée selon divers paramètres donnés aux débris et à la géométrie de la surface de rupture, basés sur les résultats de l'étude de stabilité. Cette étude montre que les processus d'instabilités récents n'affectent que les strates sédimentaires peu profondes et n'ont donc pas une grande influence sur l'élargissement du canyon. Les sédiments post-glaciaires sont déposés conformément aux vastes cicatrices de glissement, qui datent d'une période antérieure.

Bakgrund: I och med att Basel II har instiftats i Sverige så måste finansiella institutioner beräkna sin marknadsrisk på sina portföljer. Detta kan göras genom olika VaR metoder. Dessa ger dock olika uppskattningar på marknadsrisken. De finansiella instituten får använda sig av den metod som de anser reflektera marknadsrisken bäst. Det finns dock ingen metod som utsetts till standard. Syfte: Syftet med detta arbete är att jämföra olika VaR-metoders skattning av marknadsrisken utifrån verkligt utfall, för att urskilja vilken metod som är funktionsdugligast. Avgränsningar: Denna undersökning inkluderar fyra olika VaR metoder. Dessa är Historisk Simulation, Delta-Normal, RiskMetrics och GARCH(1,1). VaR metoderna kommer att undersökas på portföljer som endast består av svenska aktier noterade på Stockholmsbörsens Large-, Mid- eller Small Cap lista. Metod: Vi har konstruerat fyra olika portföljer som vi sedermera har beräknat VaR för mellan 1998-04-01 t.o.m. 2008-04-01. Dessa uppskattningar har sedermera jämförts, m.h.a. backtesting, med det verkliga utfallet för portföljerna. Utifrån detta har vi analyserat vilken form av metod som är funktionsdugligast. Resultat, slutsatser: Vi kan konstatera att ingen av de metoder som vi har undersökt är godkända enligt vår backtesting. Om vi bortser från detta så verkar RiskMetrics vara funktionsdugligast då denna metod innehar få överträdelser och uppskattar marknadsrisken på ett effektivt sätt. Detta samtidigt som RiskMetrics är stabilast under hela undersökningsperioden.

<p>Bakgrund: I och med att Basel II har instiftats i Sverige så måste finansiella institutioner beräkna sin marknadsrisk på sina portföljer. Detta kan göras genom olika VaR metoder. Dessa ger dock olika uppskattningar på marknadsrisken. De finansiella instituten får använda sig av den metod som de anser reflektera marknadsrisken bäst. Det finns dock ingen metod som utsetts till standard.</p><p>Syfte: Syftet med detta arbete är att jämföra olika VaR-metoders skattning av marknadsrisken utifrån verkligt utfall, för att urskilja vilken metod som är funktionsdugligast.</p><p>Avgränsningar: Denna undersökning inkluderar fyra olika VaR metoder. Dessa är Historisk Simulation, Delta-Normal, RiskMetrics och GARCH(1,1). VaR metoderna kommer att undersökas på portföljer som endast består av svenska aktier noterade på Stockholmsbörsens Large-, Mid- eller Small Cap lista.</p><p>Metod: Vi har konstruerat fyra olika portföljer som vi sedermera har beräknat VaR för mellan 1998-04-01 t.o.m. 2008-04-01. Dessa uppskattningar har sedermera jämförts, m.h.a. backtesting, med det verkliga utfallet för portföljerna. Utifrån detta har vi analyserat vilken form av metod som är funktionsdugligast.</p><p>Resultat, slutsatser: Vi kan konstatera att ingen av de metoder som vi har undersökt är godkända enligt vår backtesting. Om vi bortser från detta så verkar RiskMetrics vara funktionsdugligast då denna metod innehar få överträdelser och uppskattar marknadsrisken på ett effektivt sätt. Detta samtidigt som RiskMetrics är stabilast under hela undersökningsperioden.</p>

Tableau d’honneur de la Faculté des études supérieures et postdoctorales, 2007-2008. / Le phénotype d'une cellule cancéreuse résulte d'un regroupement de plusieurs caractéristiques qui constituent la forme clinique du cancer. Malgré cette diversité, il est possible de mettre en évidence un certain nombre de traits distinctifs communs à la plupart des tumeurs. Parmi ceux-ci, l'invasion tissulaire et la formation de métastases sont les causes de près de quatre-vingt dix pourcent de la mortalité attribuable au cancer. Plusieurs similitudes entre le développement embryonnaire et la cancérogenèse ont été mises en évidence. Les recherches menées durant la dernière décennie sur le rôle que jouent les oncogènes et les gènes suppresseurs de tumeurs dans le développement normal d'un organisme ont grandement amélioré notre compréhension du cancer. La différence fondamentale séparant les deux processus est qu'un organisme émane d'une suite logique d'étapes reproductibles et bien définies, alors que le développement d'une tumeur demeure complètement imprévisible. Par contre, il est clair que plusieurs de ces étapes sont partagées, par exemple la régulation du cytosquelette, de l'adhérence et de la migration de la cellule. Nous avons étudié les fonctions in vivo de deux familles de protéines kinases, soit les p21-activated kinase (PAK) et les mixed-lineage kinase (MLK), qui sont en mesure de réguler certains de ces événements. Nous avons utilisé trois approches différentes pour caractériser les fonctions de PAKl durant le développement de Xenopus. Premièrement, nous avons montré que PAKl est clivée par les caspases durant l'apoptose dans les embryons. Nous avons aussi démontré que PAKl phosphoryle directement la myosine II, et que cette activation est nécessaire et suffisante à la fragmentation cellulaire caractéristique des cellules en apoptose. Deuxièmement, nous avons découvert que PAKl induit une perte de l'adhérence cellulaire dépendante de la signalisation par le récepteur tyrosine kinase EphA4. Nous avons proposé un nouveau mécanisme pour préciser la migration cellulaire dirigée par EphA4, selon lequel l'activation du récepteur EphA4 recrute PAKl à la membrane pour séquestrer la forme active de Cdc42, et ainsi diminuer l'adhérence cellulaire. Finalement, nous avons déterminé que PAKl contrôle l'induction et la migration des cellules de la crête neurale en partie par sa régulation de l'expression de Snail et de Twist, deux instigateurs importants de la progression tumorale. En somme, par sa régulation de la dynamique du cytosquelette et de l'expression génique, PAK1 joue un rôle central dans le contrôle de l'adhérence et de la motilité des cellules. Par conséquent, PAK1 serait une cible intéressante pour la prévention de l'invasion tissulaire et de la formation de métastases. Nos travaux ont aussi mis en évidence que la fonction de MLK2, une cible potentielle de PAK1, est nécessaire à la différenciation des tubules néphrétiques dans les embryons de Xenopus. Par ailleurs, nos études ont démontré que la perte de Mlkl, de Mlkl, ou de ces deux gènes chez la souris n'entraîne aucun effet sur leur viabilité et leur fertilité, de même qu'aucun changement dans la morphologie globale des tissus dans lesquels ces deux gènes sont exprimés. Les données observées pour MLK2 chez Xenopus suggèrent que l'expression de protéines kinases restreinte à certains tissus est importante pour générer des réponses spécifiques à des signaux extracellulaires communs. De plus, l'inactivation génique des MLK suggère fortement l'existence chez les mammifères d'une redondance des fonctions associées à Mlkl et Mlk2 avec celles de l'autre membre de la famille MLK, Mlk3. / The clinical manifestation of cancer is the result of a complex set of characteristics that together create diverse cancer cell phenotypes. Despite this variety, there appear to be a number of common traits that govern the conversion of normal human cells into malignant cancer cells. Among these traits, tissue invasion and metastasis feature in the most aggressive and lethal tumors and together account for approximately ninety percent of cancer deaths. Important parallels have been made between embryonic development and the growth and spread of cancer. The last decade of research has proven that our understanding of cancer has been greatly advanced by our comprehension of the role that so-called oncogenes and tumor suppressors play in the normal development of an organism. The fondamental difference between development and cancer is that development uses regulated, reproducible cellular processes, whereas in cancer these processes become irregular, leading to an unpredictable chain of events. On the other hand, some cellular events are clearly common to both; cytoskeletal dynamics, cell adhesion and migration are such examples. We have utilized the frog embryo and mouse genetics to study the in vivo functions of two groups of protein kinases, the p21 -activated kinases (PAKs) and the mixed-lineage kinases (MLKs) that are believed to regulate these events. We used three different approaches to investigate PAK1 functions during Xenopus development. Firstly, we showed that PAK1 is activated following its cleavage by caspases during apoptosis in embryos. We demonstrated that PAK1 directly phosphorylates myosin II, and that this activation of myosin is sufficient to cause cell fragmentation, a hallmark of apoptotic cells. Secondly, we determined that PAK1 induces a loss of cell adhesion dependent of a signalling pathway downstream of the EphA4 tyrosine kinase receptor. We found that EphA4 activation sequesters active Cdc42 to down-regulate cell-cell adhesion and we proposed a novel mechanism to explain Eph-directed cell migration. Finally, we demonstrated that PAK1 regulates neural crest specification and migration, in part through its regulation of the expression of Snail and Twist, two master regulators of tumor progression. In summary, by regulating cytoskeletal dynamics and gene expression, PAK1 function appears mandatory for cell adhesion and motility, and thus would be an interesting target for the prevention of tissue invasion and metastasis. We also showed that MLK2, a potential downstream target of PAKl, is required for pronephric tubule differentiation in Xenopus embryos. On the other hand, we also revealed by gene inactivation that mice lacking Mlkl, Mlk2 or both genes are viable, fertile and do not display any obvious phenotypic defects in the tissues where these genes are expressed. While the Xenopus MLK2 data suggest that tissue restricted expression of protein kinases could play an important role in signal transduction by mediating cell-specific responses to common signals, gene inactivation in mouse strongly suggests that Mlkl and Mlk2 functions in mammals are redundant with those of the other member of the MLK family, Mlk3.

Texas’ Mass Emission Cap and Trade program is a mandatory Nitrous Oxide (NOx) abatement program for medium and large stationary sources located in the Houston-Galveston ozone non-attainment area. Effected companies are required to upgrade equipment to meet the current best achievable NOx control technology (BACT) standards or to purchase emission credits in sufficient quantity to cover the difference in emissions between existing equipment and equipment meeting the BACT standard. With over 260 participating companies, the market for emission credits is ever changing, making it difficult to evaluate whether the lowest cost decision is to upgrade equipment or to purchase NOx emission credits. Because equipment upgrades are capital investments, a well informed, rational decision can have a significant impact on the corporate balance sheet. The objective of this research is to aid the decision maker by predicting credit prices based on a Geometric Brownian Motion model based on historical NOx emission credit transactions. The predicted credit price is useful in evaluating the likelihood of the equipment upgrade option being a favorable or unfavorable decision. For the examined cases, modeled results indicate that equipment upgrade is the more cost effective option. / text

This thesis analyzes investment incentives of strategic firms in industries where either demand is uncertain, or the good produced is economically non-storable and demand fluctuates. In those industries, investment is a long run decision, whereas production has to be adjusted short-run. Prominent examples are recently liberalized utilities such as the electricity sector. Regulated monopolies have been replaced by a small number of competing firms, which often are considered to behave strategically in order to exercise market power. Whereas the regulatory regimes prior to liberalization induced generous (over-)investment choices, we observe increasing unease of experts and policy makers regarding investment incentives in liberalized electricity markets. The first three chapters of this thesis (part one) analyze total capacity choice of strategic firms prior to producing for the spot market. We first determine the equilibrium of the market game. In the remainder of the first part we analyze the interdependency of enhanced spot market competition and firms overall capacity choice. We first analyze the impact of complete elimination of market power at the spot market giving rise to marginal cost pricing. We then consider the impact of price caps at the spot market. And finally we study the impact of reduced market power at the spot markets due to forward contracting. In the second part of the thesis firms can invest into several technologies. This allows them to determine not only their total capacity but also it's precise composition. In the absence of strategic interaction, for a single regulated firm, this has already been thoroughly analyzed in the so called peak load pricing literature, which has been widely applied for electricity markets prior to liberalization. In order to accommodate for the completely changed situation after liberalization, however, we extend this framework to the case of strategically interacting firms. Based on data of the German electricity market, we then illustrate and empirically quantify our theoretical results. We determine firms’ investment choices for different market structures and quantify the impact of spot market interventions on investment decisions and welfare. This allows us to quantify the potential for the exercise of market power, in the long run, when firms’ investment decisions are taken into account.