L'estérification de Fisher est une étape importante pour développer un procédé produisant de l'éthanol à partir de biomasse gazéifiée. Cette réaction réversible permet de transformer l'acide acétique et l'alcool, produits à partir de gaz de synthèse, en acétate de méthyle ou d'éthyle. L'ester peut, par la suite, subir une hydrogénolyse pour produire de l'éthanol. Le but de ce projet est de briser l'équilibre de réaction pour obtenir une plus grande conversion dans un procédé continu. Un montage de distillation réactive a permis d'obtenir une conversion élevée pour la formation d'acétate de méthyle, c'est-à-dire de plus de 96 % avec un ratio molaire de méthanol sur acide acétique de 2. Il y avait 3 % d'acide sulfurique comme catalyseur dans l'acide acétique et le ratio de reflux était de 1,4. Lorsque ce ratio molaire était de 1, la conversion était de 89 %, mais la pureté du distillat était plus élevée, 93 % d'acétate de méthyle sur une base massique. En effet, lorsque le méthanol est utilisé en excès, il se retrouve en tête de colonne. Un bullage de gaz inerte au rebouilleur a été évalué dans le but d'emporter plus aisément l'acétate de méthyle au fur et à mesure de sa formation et ainsi obtenir un produit plus pur. Cette stratégie s'est vue inefficace, car le méthanol était emporté et ce dernier traversait plus rapidement la zone réactive, résultant en une conversion moindre. De plus, les meilleurs résultats pour la formation d'acétate d'éthyle avec la distillation réactive étaient de 77 % de conversion avec 3 % d'acide sulfurique dans l'acide acétique et un ratio de reflux de 1,45. Il y avait alors une fraction massique d'acétate d'éthyle de 0,71 dans le distillat. Un réacteur en phase vapeur a été testé avec différents catalyseurs. La résine échangeuse d'ion cationique avait tendance à se désactiver en perdant son groupement sulfonique. De plus, certaines zéolithes présentaient une activité plus élevée comme la H-Mordenite avec un ratio Si/Al de 20 et la H-Y zéolithe avec un ratio Si/Al de 31. La H-Mordenite a présenté une activité constante sur un test de 72 heures avec une conversion d'environ 83 % et une fraction massique d'acétate d'éthyle de 0,67 au condensat. La température de réaction était alors de 140 ÀC, le GHSV était de 500 et le ratio molaire éthanol sur acide acétique de un. De plus, pour favoriser la réaction vers les produits selon le principe de Le Châtelier, la zéolithe LTA 3A a été utilisée avec succès pour absorber l'eau produite. Deux lits catalytiques superposés par deux lits de dessiccant ont servi à l'expérience. Une conversion de 97 % a été atteinte avec cette zéolithe et le pourcentage d'eau dans le condensat était de 0,13%. Pour la même configuration du réacteur avec du carbure de silicium au lieu du dessiccant, la conversion était de 91 % avec 9 % d'eau dans les produits. Ces résultats ont été obtenus avec un GHSV de 250, une température de 140 ÀC et un ratio molaire éthanol sur acide acétique de 0,5.

We present new measures of the rotation curves of disc galaxies from z~0.2 to z~1.7, using deep exposures from both DEIMOS and LRIS spectrographs on the Keck telescopes in combination with multi-band imaging from the Hubble Space Telescope. We do this with a new modelling code, curvation, which has been optimised to extract the rotation velocity measurements from galaxies at intermediate and high redshift. To this end, we conduct a bulge-to-disc de-composition to allow us to de-project observed velocities to extract a model of the intrinsic rotation curve. We demonstrate the improved accuracy and precision of these measurements via a number of tests, but primarily in recovering an intrinsic scatter of the high redshift Tully-Fisher relation which is similar to that found locally. We show for the first time that the stellar mass Tully-Fisher relation is tightly in place at z~1, the normalisation of which has evolved less than 0.02±0.02 dex in stellar mass from z~1.7 to z~0.2. We do however see evidence for evolution in classic B-band Tully-Fisher relation, which is brighter at z~1 by 0.85±0.28 magnitudes than that at z~0.3. This trend is consistent with what was previously known about the evolving star-formation rates of disc galaxies. We then explore the potential drivers of these trends in the Tully-Fisher relation by estimating the baryonic and dark matter content of our galaxies. We also discover a surprising trend in the bulgeless disc galaxies at high redshift, which may be evolving differently from other rotationally supported galaxies. In the context of work which has been conducted at z~2, we discuss our results of a stellar mass Tully-Fisher relation which is strikingly similar over two-thirds of the age of the Universe.

Statistical mechanics has been applied to a wide range of systems in physics, biology, medicine and even anthropology. This theory has been recently used to model the complex biochemical processes of gene expression and regulation. In particular, genetic networks offer a large number of interesting phenomena, such as multistability and oscillatory behaviour, that can be modelled with statistical mechanics tools. In the first part of this thesis we introduce gene regulation, genetic switches, and the colonization of a spatially structured media. We also introduce statistical mechanics and some of its useful tools, such as the master equation and mean- field theories. We present simple examples that are both pedagogical and also set the basis for the study of more complicated scenarios. In the second part we consider the exclusive genetic switch, a fundamental example of genetic networks. In this system, two proteins compete to regulate each other's dynamics. We characterize the switch by solving the stationary state in different limits of the protein binding and unbinding rates. We perform a study of the bistability of the system by examining its probability distribution, and by applying information theory techniques. We then present several versions of a mean field theory that offers further information about the switch. Finally, we compute the stationary probability distribution with an exact perturbative approach in the unbinding parameter, obtaining a valid result for a wide range of parameters values. The techniques used for this calculation are successfully applied to other switches. The topic studied in the third part of the thesis is the propagation of a trait inside an expanding population. This trait may represent resistance to an antibiotic or being infected with a certain virus. Although our model accounts for different examples in the genetic context, it is also very useful for the general study of a trait propagating in a population. We compute the speed of expansion and the stationary population densities for the invasion of an established and an expanding population, finding non-trivial criteria for speed selection and interesting speed transitions. The obtained formulae for the different wave speeds show excellent agreement with the results provided by simulations. Moreover, we are able to obtain the value of the speeds through a detailed analysis of the populations, and establish the requirements for our equations to present speed transitions. We finally apply our model to the propagation in a position-dependent fitness landscape. In this situation, the growth rate or the maximum concentration depends on the position. The amplitudes and speeds of the waves are again successfully predicted in every case.

Thesis (PhD (Statistics and Actuarial Science))—University of Stellenbosch, 2008. / The problem of variable selection in binary kernel classification is addressed in this thesis. Kernel methods are fairly recent additions to the statistical toolbox, having originated approximately two decades ago in machine learning and artificial intelligence. These methods are growing in popularity and are already frequently applied in regression and classification problems. Variable selection is an important step in many statistical applications. Thereby a better understanding of the problem being investigated is achieved, and subsequent analyses of the data frequently yield more accurate results if irrelevant variables have been eliminated. It is therefore obviously important to investigate aspects of variable selection for kernel methods. Chapter 2 of the thesis is an introduction to the main part presented in Chapters 3 to 6. In Chapter 2 some general background material on kernel methods is firstly provided, along with an introduction to variable selection. Empirical evidence is presented substantiating the claim that variable selection is a worthwhile enterprise in kernel classification problems. Several aspects which complicate variable selection in kernel methods are discussed. An important property of kernel methods is that the original data are effectively transformed before a classification algorithm is applied to it. The space in which the original data reside is called input space, while the transformed data occupy part of a feature space. In Chapter 3 we investigate whether variable selection should be performed in input space or rather in feature space. A new approach to selection, so-called feature-toinput space selection, is also proposed. This approach has the attractive property of combining information generated in feature space with easy interpretation in input space. An empirical study reveals that effective variable selection requires utilisation of at least some information from feature space. Having confirmed in Chapter 3 that variable selection should preferably be done in feature space, the focus in Chapter 4 is on two classes of selecion criteria operating in feature space: criteria which are independent of the specific kernel classification algorithm and criteria which depend on this algorithm. In this regard we concentrate on two kernel classifiers, viz. support vector machines and kernel Fisher discriminant analysis, both of which are described in some detail in Chapter 4. The chapter closes with a simulation study showing that two of the algorithm-independent criteria are very competitive with the more sophisticated algorithm-dependent ones. In Chapter 5 we incorporate a specific strategy for searching through the space of variable subsets into our investigation. Evidence in the literature strongly suggests that backward elimination is preferable to forward selection in this regard, and we therefore focus on recursive feature elimination. Zero- and first-order forms of the new selection criteria proposed earlier in the thesis are presented for use in recursive feature elimination and their properties are investigated in a numerical study. It is found that some of the simpler zeroorder criteria perform better than the more complicated first-order ones. Up to the end of Chapter 5 it is assumed that the number of variables to select is known. We do away with this restriction in Chapter 6 and propose a simple criterion which uses the data to identify this number when a support vector machine is used. The proposed criterion is investigated in a simulation study and compared to cross-validation, which can also be used for this purpose. We find that the proposed criterion performs well. The thesis concludes in Chapter 7 with a summary and several discussions for further research.

Thesis (PhD (Statistics and Actuarial Science))—Stellenbosch University, 2008. / Kernel Fisher discriminant analysis (KFDA) is a kernel-based technique that can be used to classify observations of unknown origin into predefined groups. Basically, KFDA can be viewed as a non-linear extension of Fisher’s linear discriminant analysis (FLDA). In this thesis we give a detailed explanation how FLDA is generalized to obtain KFDA. We also discuss two methods that are related to KFDA. Our focus is on binary classification. The influence of atypical cases in discriminant analysis has been investigated by many researchers. In this thesis we investigate the influence of atypical cases on certain aspects of KFDA. One important aspect of interest is the generalization performance of the KFD classifier. Several other aspects are also investigated with the aim of developing criteria that can be used to identify cases that are detrimental to the KFD generalization performance. The investigation is done via a Monte Carlo simulation study. The output of KFDA can also be used to obtain the posterior probabilities of belonging to the two classes. In this thesis we discuss two approaches to estimate posterior probabilities in KFDA. Two new KFD classifiers are also derived which use these probabilities to classify observations, and their performance is compared to that of the original KFD classifier. The main objective of this thesis is to develop criteria which can be used to identify cases that are detrimental to the KFD generalization performance. Nine such criteria are proposed and their merit investigated in a Monte Carlo simulation study as well as on real-world data sets. Evaluating the criteria on a leave-one-out basis poses a computational challenge, especially for large data sets. In this thesis we also propose using the smallest enclosing hypersphere as a filter, to reduce the amount of computations. The effectiveness of the filter is tested in a Monte Carlo simulation study as well as on real-world data sets.

Information theoretical measures are used to design, from first principles, an objective function that can drive a learning machine process to a solution that is robust to perturbations in parameters. Full analytic derivations are given and tested with computational examples showing that indeed the procedure is successful. The final solution, implemented by a robust learning machine, expresses a balance between Shannon differential entropy and Fisher information. This is also surprising in being an analytical relation, given the purely numerical operations of the learning machine.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Melanoma differentiation associated gene-7 (mda-7) was discovered in the Fisher laboratory by subtraction hybridization of temporally spaced subtracted cDNA libraries prepared from terminally differentiated human melanoma cells treated with human fibroblast interferon (IFN-β) and the protein kinase C activator mezerein (MEZ), an approach called ‘differentiation induction subtraction hybridization’ (DISH). mda-7 is located in human chromosome 1q32–33 and based on sequence homology, chromosomal localization, and its functional properties, the mda-7 gene is now classified as a member of the IL-10 family of cytokines and named IL-24. The mda-7/IL-24 cDNA encodes a protein of 206-amino acids with a predicted size of ~24-kDa, which contains an interleukin (IL)-10 signature motif at amino acids 101–121 (SDAESCYLVHTLLEFYLKTVF) shared by other members of the IL-10 family of cytokines. Sequence analysis revealed the presence of a 49-amino acid signal peptide suggesting that the molecule could be cleaved and secreted. Expression of MDA-7/IL-24 protein was detected in cells of the immune system (mainly by expression in tissues associated with the immune system, such as spleen, thymus and PBMC) and normal human melanocytes. Of interest, a progressive loss of MDA-7/IL-24 expression during melanoma progression suggests an inverse relationship between MDA-7/IL-24 expression and the evolution of melanocytes to various stages of melanoma. mda-7/IL-24 induces growth suppression in human melanoma and other cancer cells, without affecting normal cells. Subsequent studies provided consistent evidence that ectopic expression of mda-7/IL-24 employing a replication incompetent adenovirus (Ad.mda-7) resulted in apoptosis induction and cell death in a wide variety of solid tumors including melanoma, malignant glioma, carcinomas of the breast, kidney, cervix, colorectum , liver, lung, ovary and prostate sparing normal cellular counterparts, i.e., such as normal melanocytes, astrocytes, fibroblasts, and mesothelial and epithelial cells. The in vitro antitumor activity of mda-7/IL-24 readily translated into the in vivo situation in animal models containing human breast, prostate, lung and colorectal carcinomas and in malignant glioma xenografts. Moreover, the ability of mda-7/IL-24 to induce a potent “bystander cancer-specific killing effect” provides an unprecedented opportunity to use this molecule to target for destruction not only primary tumors, but also metastases. Based on its profound cancer-selective tropism, substantiated by in vivo human xenograft studies in nude mice, mda-7/IL-24 (administered as Ad.mda-7) was evaluated in a Phase I clinical trial in patients with melanomas and solid cancers. These studies document that mda-7/IL-24 is well tolerated and demonstrates evidence of significant (44%) clinical activity. This review focuses on the recent enhancements in our understanding of the mode of action of mda-7/IL-24 and its potential applications as a unique and promising effective cytokine-based gene therapy for human cancers. The first chapter explored the efficacy of a tropism-modified Ad-based cancer gene therapy approach for eradicating low CAR colorectal cancer cells. We show that in low CAR human colorectal cancer cells (RKO), a recombinant Ad.5/3 virus delivering mda-7/IL-24 (Ad.5/3-mda-7) is more efficient than Ad.5 delivering mda-7 (Ad.5-mda-7) in expressing MDA-7/IL-24 protein, inducing cancer-specific apoptosis and inhibiting in vivo tumor growth in a nude mouse xenograft model. Additionally, our in vitro and in vivo data confirms that BI-97C1 (Sabutoclax) profoundly sensitizes mda-7/IL-24 mediated toxicity in colorectal cancer. Thus, Ad.5/3-mda-7, alone and/or in combination with BI-97C1 (Sabutoclax), might represent an improved and more effective therapeutic approach for colorectal and other cancers. In view of the essential roles of anti-apoptotic Bcl-2 family proteins in tumorigenesis and chemoresistance, efforts are focused on developing small molecule inhibitors of Bcl-2 family proteins as potential therapeutics for cancer. Unfortunately, due to the unique structure of Mcl-1 as compared with Bcl-2 and Bcl-xL, currently employed inhibitors, such as ABT-737 or its clinical counterpart, ABT-263, display limited affinity for Mcl-1. Using nuclear magnetic resonance (NMR) binding assays and computational docking studies, we have recently identified a series of new Apogossypol derivatives, compound 3 (BI-79D10) and compound 11 (BI-97C1), with pan-Bcl-2- inhibitory potency. BI-79D10 binds to Bcl- xL, Bcl-2, and Mcl-1 with IC50 values of 190, 360, and 520 nmol/L, respectively. BI-97C1 (Sabutoclax) is an optically pure individual Apogossypol derivative that retains all the properties of BI-79D10 along with superior in vitro and in vivo efficacy. Because Mcl-1 is over-expressed in the majority of PCs, we hypothesized that suppressing Mcl-1 by treating human PC cells with BI-97C1 (Sabutoclax) would sensitize them to mda-7/IL-24-mediated cytotoxicity. The second chapter study highlights the noteworthy potential of a combinatorial approach involving mda-7/IL-24, a broad-acting anticancer gene, and BI-97C1 (Sabutoclax), which targets Mcl-1, to sensitize PC to mda-7/IL-24-mediated cytotoxicity, thereby enhancing therapeutic efficacy. Our data suggests that treatment with the combination regimen of mda-7/IL-24 and BI-97C1 (Sabutoclax) induces autophagy that facilitates apoptosis in association with up regulation of NOXA, accumulation of Bim, and activation of Bax and Bak. Treatment with mda-7/IL-24 and BI-97C1 (Sabutoclax) inhibited the growth of PC xenografts and suppressed PC development in an immunocompetent transgenic mouse model of PC. The third chapter study explored the efficacy of a tropism-modified CRCA cancer gene therapy approach for eradicating low CAR prostate cancer cells. We showed that in low CAR PC3 cells Ad.5/3-CTV is more efficient than Ad.5-CTV in delivering transgene (mda-7/IL-24), infecting tumor cells, expressing MDA-7/IL-24 protein, inducing cancer-specific apoptosis, inhibiting in vivo tumor growth and exerting an antitumor ‘bystander’ effect in a nude mouse human prostate cancer xenograft and suppressed PC development in an immunocompetent transgenic mouse model of PC model.

Throughout the course of this thesis, we give an introduction to optimal control theory and its necessary conditions, prove Pontryagin's Maximum Principle, and present the life-cycle saving under uncertain lifetime optimal control problem. We present a very involved sensitivity analysis that determines how a change in the initial wealth, discount factor, or relative risk aversion coefficient may affect the model the terminal depletion of wealth time, optimal consumption path, and optimal accumulation of wealth path. Through simulation of the life-cycle saving under uncertain lifetime model, we are not only able to present the model dynamics through time, but also to demonstrate the feasibility of the model.

Thesis advisor: Bonnie Jefferson / This thesis analyzes the discourse surrounding two of the most celebrated professional athletes in the present generation. Kobe Bryant and LeBron James are two highly talented basketball players who have both been hailed within the world of sports ever since they left high school and entered directly into the NBA. This study argues that the media has presented the careers of both Bryant and James in the form of carefully constructed and familiar narratives. The analysis incorporates concepts drawn from Walter R. Fisher, Seymour Chatman, Karyn and Donald Rybacki and Kenneth Burke, in order scrutinize the narrative elements existing within specific artifacts presented by the media. The analysis demonstrates how the selected artifacts uphold plotlines that are already recognizable to the audience. Through influential rhetorical devices, the media frames the careers of Bryant and James so that the two men are featured as the mythological heroes of their tales. This study reveals the power of framing a message as an identifiable narrative as well as the implications the construction has for both the athletes and the audience. / Thesis (BA) — Boston College, 2010. / Submitted to: Boston College. College of Arts and Sciences. / Discipline: Communication Honors Program. / Discipline: Communication.