Le resvératrol, stilbène produit naturellement dans les grappes de raisin, possède un pouvoir anti-tumoral caractérisé par ses effets pléïotropiques. La molécule s’est ainsi montrée capable d’inhiber la cancérogenèse en agissant à de multiples niveaux. Son incidence sur la croissance tumorale in vitro est généralement associée à une perturbation dans le cycle cellulaire et à une induction d’apoptose. De nombreuses molécules dérivées du resvératrol se sont montrées également efficaces pour moduler la croissance tumorale. Nous avons entrepris au cours de ces travaux de thèse d’étudier les activités biologiques de stilbènes issus d’une production de culture cellulaire de vigne. Les résultats de cette étude mettent en évidence une réduction de la croissance tumorale, dans un modèle d’étude in vitro de mélanome humain, pour le resvératrol et quatre oligomères de resvératrol bioproduits. L’étude plus approfondie de trois oligomères de resvératrol a permis de révéler des différences notables par rapport entre les effets du resvératrol et ses dérivés. En effet, bien que l’inhibition de la viabilité cellulaire des lignées cancéreuses s’accompagne d’une perturbation du cycle cellulaire pour le resvératrol et ses dérivés, elle se traduit différemment selon les composés. L’action du resvératrol sur le cycle cellulaire conduit à une nette augmentation des cellules en phase S, tandis que les oligomères de resvératrol n’induisent pas d’accumulation cellulaire franche dans une phase du cycle. De plus les oligomères de resvératrol présentent des capacités anti-invasives et anti-migratoires qui ne sont pas retrouvées chez le resvératrol. Enfin l’étude sur des fibroblastes normaux d’adultes (FNA) souligne que les dérivés du resvératrol diminuent la viabilité de cellules normales et cancéreuses à des concentrations similaires contrairement au resvératrol dont l’action est nettement inférieure sur les cellules saines. Cependant le pouvoir cytotoxique du resvératrol et de ses dérivés reste beaucoup plus faible chez les cellules normales. / Resveratrol, a natural stilbene found in the grapevine, exhibits pleiotropic antitumor activities. This molecule has been shown to inhibit the cancerogenesis processes at different levels. Its impact in tumor growth, in vitro, is generally associate with a disruption in cell cycle and an apoptosis induction. Various resveratrol derivatives appear also efficient to modulate tumor growth. We have studied, during this work, biological activities of stilbenes produced by grapevine cell cutures. Our results show a decrease in melanoma human growth, in vitro, for bioproduced resveratrol oligomers as well as resveratrol itself. Our experiments underline distinct effects between resveratrol and its derivatives. Indeed, all compounds disrupt cell cycle, but resveratrol induces S-phase arrest while resveratrol oligomers failed to induce a clearly phase arrest. Moreover, only resveratrol oligomers only have the ability to counteract invasive and migratory properties of melanoma cells. Finally, experiments performed in normal human fibroblasts study show that resveratrol oligomers present a similar potential to reduce viability of cancer as well as normal cells contrary to resveratrol which is less efficient in normal cells. However the cytotoxicity of resveratrol and its derivatives is significantly reduced on normal cells.

Context. Real time rendering has many challenges to overcome, one of them being character realism. One way to move towards realism is to use wrinkle maps. Although already used in several games, there might be room for improvement, common practice suggests using two wrinkle maps, however, if this number can be reduced both texture usage and workload might be reduced as well. Objectives. To determine whether or not it is possible to reduce the number of wrinkle maps from two to one without having any significant impact on the perceived realism of a character. Methods. After a base character model was created, a setup in Maya were made so that dynamic wrinkles could be displayed on the character using both one and two wrinkle maps. The face were animated and rendered, displaying emotions using both techniques. A two-alternative forced choice experiment was then conducted where the participants selected which implementation displaying the same facial expression and having the same lighting condition they perceived as most realistic. Results. Results showed that some facial expressions had more of an impact of the perceived realism than others, favoring two wrinkle maps in every case where there was a significant difference. The expressions with the most impact were the ones that required different kinds of wrinkles at the same area of the face, such as the forehead, where one variant of wrinkles run at a more vertical manner and the other variant runs horizontally along the forehead. Conclusions. Using one wrinkle map can not fully replicate the effect of using two when it comes to realism. The difference on the implementations are dependant on the expression being displayed.

Dagens spel har väldigt väl utvecklade karaktärer men hur skapar man en karaktär egentligen? Denna rapport kommer att handla om processen för att skapa en spelkaraktär, man kommer att ta en titt på vilka verktyg och program som kan användas sig av när man skapar en spelkaraktär. Man kommer även skapa en egen karaktär och använda sig utav några utvalda program för att skapa den och kommer även undersöka lite om de program som har använts i skapandet av den karaktär som kommer skapas för att få en bättre förståelse för hur man kan göra när man skapar en karaktär och kommer i slutet gå igenom resultatet av arbetet på denna karaktär och vad man har kommit fram till. Kraven på en designer är väldigt höga och processen för att skapa en spelkaraktär kan se ut på olika sätt så i denna rapport prövar vi en process.

Geógrafo / La búsqueda de la localización óptima para infraestructuras culturales especializadas, como es el caso de los centros culturales, motivó el desarrollo de esta investigación. En dicha localización se entrecruzan consideraciones de índole presupuestaria y administrativa con aspectos relativos al entorno de su emplazamiento. Nos remite a conceptos como centralidad y proximidad, con respecto a otras infraestructuras o a medios de transporte; ¿cuál es el peso que juegan elementos característicos del mercado, como oferta y demanda?, en este caso, por bienes y servicios culturales. Teniendo como antecedente la existencia de factores clásicos de localización de actividades industriales y de servicios (distancias, costos de transporte, demanda, puntos de oferta, etc.), se intentó identificar tanto la utilidad de ellos como la posibilidad de incluir nuevos elementos en la búsqueda del mejor lugar de localización para centros culturales públicos de proximidad (centros culturales de alcance local). La preparación de los diversos componentes para la localización y la posterior prueba del modelo diseñado, se realizó a partir de un Proceso Analítico Jerárquico (AHP). Finalmente, si bien esta memoria se centra en criterios de tipo físico, de infraestructura y de accesibilidad, se quiere declarar que cada uno de los elementos mencionados adquiere sentido cuando se incorpora a la comunidad que se espera atender, sus intereses, sueños y formas de interrelación. El centro cultural logra cumplir con su misión y justifica su existencia en un lugar determinado, cuando es vivido y recreado por sus vecinos y habitantes.

Les phénomènes de contact impliquant des corps déformables abondent dans l'industrie, notamment dans les structures mécaniques. En raison de leur complexité intrinsèque, les phénomènes de contact sont modélisés à l'aide de problèmes aux limites fortement non linéaires. De ce fait, la modélisation de ces phénomènes pose plusieurs difficultés mathématiques. Dans cette thèse, nous nous intéressons à la modélisation, l'analyse variationnelle et l'analyse numérique de problèmes de contact glissant intervenant en mécanique des solides, pour des matériaux élastiques, viscoélastiques et viscoplastiques. La première partie de cette thèse rappelle quelques résultats préliminaires, notamment des outils mathématiques et mécaniques nécessaires pour réaliser la suite de ce travail. La deuxième partie est consacrée à l'étude de quelques problèmes de contact glissant sous diverses conditions de contact et frottement. Pour chacun de ces problèmes, nous introduisons les formulations fortes et des formulations variationnelles. Ensuite, nous obtenons des résultats d'existence et d'unicité des solutions faibles, sous certaines hypothèses de petitesse, ainsi que des résultats de convergence. Enfin, nous proposons une approximation numérique de certains problèmes de contact à l'aide de schémas discrétisés. Pour ces schémas, nous obtenons des résultats d'estimation de l'erreur. / Contact phenomena involving deformable bodies abound in industry, especially in mechanical structures. Due to their intrinsic complexity, contact phenomena are modelled with strongly nonlinear boundary problems. For this reason, the modelling of these phenomena gives rise to various mathematical difficulties. In this thesis, we are interested in the modelling, the variational analysis and the numerical analysis of problems of sliding contact in solid mechanics for elastic, viscoelastic and viscoplastic materials. The first part of this thesis concerns some preliminary results, in particular the mathematical and mechanical tools necessary to carry out the continuation of this work. The second part is devoted to the study of some problems of sliding contact under various conditions of contact and friction. For each of these problems, we introduce strong formulations and variational formulations. Then, we obtain results of existence and uniqueness of the weak solutions, under smallness assumptions, as well as results of convergence. Finally, we propose a numerical approximation of some contact problems based on the etudy of discretized schemes. For these schemes we obtain error estimates results.

Investigation of the thermal quenching of photoluminescence (PL) in semiconductors provides valuable information on identity and characteristics of point defects in these materials, which helps to better understand and improve the properties of semiconductor materials and devices. Abrupt and tunable thermal quenching (ATQ) of PL is a relatively new phenomenon with an unusual behavior of PL. This mechanism was able to explain what a traditional model failed to explain. Usually, in traditional model used to explain “normal” quenching, the slope of PL quenching in the Arrhenius plot determines the ionization energy of the defect causing the PL band. However, in abrupt quenching when the intensity of PL decreases by several orders of magnitude within a small range of temperature, the slope in the Arrhenius plot has no relation to the ionization energy of any defect. It is not known a priori if the thermal quenching of a particular PL band is normal or abrupt and tunable. Studying new cases of unusual thermal quenching, classifying and explaining them helps to predict new cases and understand deeper the ATQ mechanism of PL thermal quenching. Very few examples of abrupt and tunable quenching of PL in semiconductors can be found in literature. The abrupt and tunable thermal quenching, reported here for the first time for high-resistivity ZnO, provides an evidence to settle the dispute concerning the energy position of the LiZn acceptor. In high-resistivity GaN samples, the common PL bands related to defects are the yellow luminescence (YL) band and a broad band in the blue spectral region (BL2). In this work, we report for the first time the observation of abrupt and tunable thermal quenching of the YL band in GaN. The activation energies for the YL and BL2 bands calculated through the new mechanism show agreement with the reported values. From this study we predict that the ATQ phenomenon is quite common for high-resistivity semiconductors.

The problem of estimating mean and covariances of a multivariate normal distributedrandom vector has been studied in many forms. This thesis focuses on the estimatorsproposed in [15] for a banded covariance structure with m-dependence. It presents theprevious results of the estimator and rewrites the estimator when m = 1, thus makingit easier to analyze. This leads to an adjustment, and a proposition for an unbiasedestimator can be presented. A new and easier proof of consistency is then presented.This theory is later generalized into a general linear model where the correspondingtheorems and propositions are made to establish unbiasedness and consistency. In thelast chapter some simulations with the previous and new estimator verifies that thetheoretical results indeed makes an impact.

Master of Public Health - MPH / A matched case-control study was carried out to identify the factors associated with the nutritional status of children in Livingstone district, Zambia. A trained field researcher collected data on demographic, feeding practices and socio-economic factors in matched groups of underweight children and normal weight children from the mothers/caregivers of the children. Interviews, using a structured pre-tested questionnaire, were used to obtain the data. All children aged from six to fifty-nine months admitted at Livingstone General Hospital during the months of October to December, 2003 with a weight for age below –2SD from the median of the reference population formed the underweight children (cases). The total was 47 children. The cases were then matched according to sex and age to 47 normal weight children attending the under five clinic at Livingstone General Hospital during the same months, with a weight for age above –2SD from the median of the reference population (controls). The final sample was 94 children. Weights were recorded using the same scale in the outpatient’s department prior to admission and during the under five clinic session. Interviews then took place in a separate room after the clinic nurse had attended them. A separate visit was then made to the homes of the children on a different day. Odds ratios and 95% confidence intervals were used to measure strengths of associations. Results revealed that the basic causes of underweight were the poor economic state of the country, unfavorable policies and insufficient government support in the areas of health, education, agriculture, housing and employment. Underlying causes were inadequate access to food, inadequate care of children, poor access to health services and unhealthy living environments, while immediate causes were poor food intake and disease. Low educational and literacy levels of the mothers/caregivers, unemployment and lack of sufficient finances to access basic necessities such as food, housing and health contributed to underweight. / South Africa

>Magister Scientiae - MSc / Risk factors such as intrauterine and vaginal infection put pregnant women at risk for delivering preterm. Bacterial vaginosis (BV) is a polymicrobial clinical syndrome commonly diagnosed in women of reproductive age, with women of African descent with low socioeconomic status and previous preterm delivery at high risk. Although frequently isolated from healthy women, Gardnerella vaginalis has been most frequently associated with BV. There is limited data available on the prevalence of BV in Southern Africa; therefore, we embarked on a study to determine the prevalence of BV and G. vaginalis in predominantly black communities in the Western Cape, in order to establish the role of G. vaginalis in BV. Women attending various Maternity and Obstetrics units (MOU) in the Cape Peninsula with and without a history of pre-term delivery (PTD) were invited to participate in the study. Several factors were statistically associated with pregnancy history, including location of study population, parity, smoking and presence of clinical symptoms. The presence of G. vaginalis was determined by culture in 51.7% of the preterm delivery group (PTDG) and 44% of the full-term delivery group (FTDG) women. BV was detected in 31.13% of PTDG and 23.67% of FTDG by Gram stained analysis according to Nugent scoring criteria, with age and HIV status posing as risk factors. When comparing PTDG and FTDG for an association between the presence of G. vaginalis and BV, a stronger association was observed in the PTDG but it was not statistically significant. In both PTDG and FTDG, G. vaginalis was isolated significantly more often in women diagnosed with BV at 24.5% (p < 0.05). Antibiogram studies revealed both Metronidazole and Clindamycin resistant strains of G. vaginalis. G. vaginalis Biotype 7 is specifically associated with BV, while Biotype 2 appears to be associated with BV in women with a history of PTD. Accuracy of diagnostic tools were tested and it was determined that Nugent scoring is more sensitive in diagnosing BV (76.04%), but culture for G. vaginalis is more specific (83.21%). Although this study was limited in that we were unable to follow-up pregnancy outcomes, we were able to confirm the perceived role of G. vaginalis in BV. / South Africa

Philosophiae Doctor - PhD / Missing data are common in survey data sets. Enrolled subjects do not often have data recorded for all variables of interest. The inappropriate handling of missing data may lead to bias in the estimates and incorrect inferences. Therefore, special attention is needed when analysing incomplete data. The multivariate normal imputation (MVNI) and the multiple imputation by chained equations (MICE) have emerged as the best techniques to impute or fills in missing data. The former assumes a normal distribution of the variables in the imputation model, but can also handle missing data whose distributions are not normal. The latter fills in missing values taking into account the distributional form of the variables to be imputed. The aim of this study was to determine the performance of these methods when data are missing at random (MAR) or completely at random (MCAR) on unordered or nominal categorical variables treated as predictors or response variables in the regression models. Both dichotomous and polytomous variables were considered in the analysis. The baseline data used was the 2007 Demographic and Health Survey (DHS) from the Democratic Republic of Congo. The analysis model of interest was the logistic regression model of the woman’s contraceptive method use status on her marital status, controlling or not for other covariates (continuous, nominal and ordinal). Based on the data set with missing values, data sets with missing at random and missing completely at random observations on either the covariates or response variables measured on nominal scale were first simulated, and then used for imputation purposes. Under MVNI method, unordered categorical variables were first dichotomised, and then K − 1 (where K is the number of levels of the categorical variable of interest) dichotomised variables were included in the imputation model, leaving the other category as a reference. These variables were imputed as continuous variables using a linear regression model. Imputation with MICE considered the distributional form of each variable to be imputed. That is, imputations were drawn using binary and multinomial logistic regressions for dichotomous and polytomous variables respectively. The performance of these methods was evaluated in terms of bias and standard errors in regression coefficients that were estimated to determine the association between the woman’s contraceptive methods use status and her marital status, controlling or not for other types of variables. The analysis was done assuming that the sample was not weighted fi then the sample weight was taken into account to assess whether the sample design would affect the performance of the multiple imputation methods of interest, namely MVNI and MICE. As expected, the results showed that for all the models, MVNI and MICE produced less biased smaller standard errors than the case deletion (CD) method, which discards items with missing values from the analysis. Moreover, it was found that when data were missing (MCAR or MAR) on the nominal variables that were treated as predictors in the regression model, MVNI reduced bias in the regression coefficients and standard errors compared to MICE, for both unweighted and weighted data sets. On the other hand, the results indicated that MICE outperforms MVNI when data were missing on the response variables, either the binary or polytomous. Furthermore, it was noted that the sample design (sample weights), the rates of missingness and the missing data mechanisms (MCAR or MAR) did not affect the behaviour of the multiple imputation methods that were considered in this study. Thus, based on these results, it can be concluded that when missing values are present on the outcome variables measured on a nominal scale in regression models, the distributional form of the variable with missing values should be taken into account. When these variables are used as predictors (with missing observations), the parametric imputation approach (MVNI) would be a better option than MICE.