Дисертації з теми "GBM approach"

Glioblastoma multiforme is the most common and lethal primary brain tumour in adults. Its aggressive and invasive phenotype makes it resistant to current standards of care, with a patient median survival following treatment of only 14 months. Potent and safe therapeutics are necessary to improve patient prognosis. Globally, efforts are being made in immunotherapies to combat such deleterious tumours. Preliminary work in the Stojdl lab has developed a novel oncolytic virus platform for brain cancer therapy that is non-toxic and exhibits potent anti-tumour efficacy. This platform is based on the rhabdovirus Farmington, identified for its potent oncolytic properties and engineering malleability. Herein, we begin to show our capability to discover and vaccinate against immunogenic neoepitopes derived from a mouse cancer mutanome. Engineering Farmington virus to express neoepitopes, allows for robust tumour specific immune proliferation following a prime vaccination. Overcoming problems of targeting self-antigen and antigen loss variants, a multi-neoepitope vaccine, presented here, is one of many alternative approaches to help combat cancer resistance. Despite achieving robust anti-tumour immunity by vaccination, selectivity of the tumour microenvironment remains an enormous challenge. Cumulative efforts in immunotherapy research will help drive novel therapeutics, like Farmington, into clinic and, ultimately, improve patient’s prognosis and quality of life.

Uma das principais áreas de pesquisa em Biologia de Sistemas refere-se à descoberta de redes biológicas a partir de conjuntos de dados de microarrays. Estas redes consistem de um grande número de genes cujos níveis de expressão afetam os outros genes de vários modos. Nesta tese, apresenta-se uma nova maneira de analisar os conjuntos de dados de microarrays, com base nos diferentes tipos de ciclos encontrados entre os genes das redes de co-expressão construídas com dados quantificados obtidos a partir dos microarrays. A entrada do método de análise é formada pelos dados brutos, um conjunto de genes de interesse (por exemplo, genes de uma via conhecida) e uma função (ativador ou inibidor) destes genes. A saída do método é um conjunto de ciclos. Um ciclo é um caminho fechado com todos os vértices (exceto o primeiro e o último) distintos. Graças à nova forma de encontrar relações entre os genes, é possível uma interpretação mais robusta das correlações dos genes, porque os ciclos estão associados a mecanismos de feedback, que são muito comuns em redes biológicas. A hipótese é que feedbacks negativos permitem encontrar relações entre os genes que podem ajudar a explicar a estabilidade do processo regulatório dentro da célula. Ciclos de feedback positivo, por outro lado, podem mostrar a quantidade de desequilíbrio de uma determinada célula em um determinado momento. A análise baseada em ciclos permite identificar a relação estequiométrica entre os genes da rede. Esta metodologia proporciona uma melhor compreensão da biologia do tumor. Portanto, as principais contribuições desta tese são: (i) um novo método de análise baseada em ciclos; (ii) um novo método de classificação; (iii) e, finalmente, aplicação dos métodos e a obtenção de resultados práticos. A metodologia proposta foi utilizada para analisar os genes de quatro redes fortemente relacionadas com o câncer - apoptose, glicólise, ciclo celular e NF B - em tecidos do tipo mais agressivo de tumor cerebral (Gliobastoma multiforme - GBM) e em tecidos cerebrais saudáveis. A maioria dos pacientes com GBM morrem em menos de um ano, essencialmente nenhum paciente tem sobrevivência a longo prazo, por isso estes tumores têm atraído atenção significativa. Os principais resultados nesta tese mostram que a relação estequiométrica entre genes envolvidos na apoptose, glicólise, ciclo celular e NF B está desequilibrada em amostras de GBM em comparação as amostras de controle. Este desequilíbrio pode ser medido e explicado pela identificação de um percentual maior de ciclos positivos nas redes das primeiras amostras. Esta conclusão ajuda a entender mais sobre a biologia deste tipo de tumor. O método de classificação baseado no ciclo proposto obteve as mesmas métricas de desempenho como uma rede neural, um método clássico de classificação. No entanto, o método proposto tem uma vantagem significativa em relação às redes neurais. O método de classificação proposto não só classifica as amostras, fornecendo diagnóstico, mas também explica porque as amostras foram classificadas de uma certa maneira em termos dos mecanismos de feedback que estão presentes/ausentes. Desta forma, o método fornece dicas para bioquímicos sobre possíveis experiências laboratoriais, bem como sobre potenciais genes alvo de terapias.
One of the main research areas in Systems Biology concerns the discovery of biological networks from microarray datasets. These networks consist of a great number of genes whose expression levels affect each other in various ways. We present a new way of analyzing microarray datasets, based on the different kind of cycles found among genes of the co-expression networks constructed using quantized data obtained from the microarrays. The input of the analysis method is formed by raw data, a set of interest genes (for example, genes from a known pathway) and a function (activator or inhibitor) of these genes. The output of the method is a set of cycles. A cycle is a closed walk, in which all vertices (except the first and last) are distinct. Thanks to the new way of finding relations among genes, a more robust interpretation of gene correlations is possible, because cycles are associated with feedback mechanisms that are very common in biological networks. Our hypothesis is that negative feedbacks allow finding relations among genes that may help explaining the stability of the regulatory process within the cell. Positive feedback cycles, on the other hand, may show the amount of imbalance of a certain cell in a given time. The cycle-based analysis allows identifying the stoichiometric relationship between the genes of the network. This methodology provides a better understanding of the biology of tumors. As a consequence, it may enable the development of more effective treatment therapies. Furthermore, cycles help differentiate, measure and explain the phenomena identified in healthy and diseased tissues. Cycles may also be used as a new method for classification of samples of a microarray (cancer diagnosis). Compared to other classification methods, cycle-based classification provides a richer explanation of the proposed classification, that can give hints on the possible therapies. Therefore, the main contributions of this thesis are: (i) a new cycle-based analysis method; (ii) a new microarray samples classification method; (iii) and, finally, application and achievement of practical results. We use the proposed methodology to analyze the genes of four networks closely related with cancer - apoptosis, glucolysis, cell cycle and NF B - in tissues of the most aggressive type of brain tumor (Gliobastoma multiforme – GBM) and in healthy tissues. Because most patients with GBMs die in less than a year, and essentially no patient has long-term survival, these tumors have drawn significant attention. Our main results show that the stoichiometric relationship between genes involved in apoptosis, glucolysis, cell cycle and NF B pathways is unbalanced in GBM samples versus control samples. This dysregulation can be measured and explained by the identification of a higher percentage of positive cycles in these networks. This conclusion helps to understand more about the biology of this tumor type. The proposed cycle-based classification method achieved the same performance metrics as a neural network, a classical classification method. However, our method has a significant advantage with respect to neural networks. The proposed classification method not only classifies samples, providing diagnosis, but also explains why samples were classified in a certain way in terms of the feedback mechanisms that are present/absent. This way, the method provides hints to biochemists about possible laboratory experiments, as well as on potential drug target genes.

In estimating econometric time series models, it is assumed that the parameters remain constant over the period examined. This assumption may not always be valid when using data which span an extended period, as the underlying relationships between the variables in these models are exposed to various exogenous shifts. It is therefore imperative to examine the stability of models as failure to identify any changes could result in wrong predictions or inappropriate policy recommendations. This research proposes a method of estimating the location of break points in linear econometric models with endogenous regressors, estimated using Generalised Method of Moments (GMM). The proposed estimation method is based on Wald, Lagrange Multiplier and Difference type test statistics of parameter variation. In this study, the equation which sets out the relationship between the endogenous regressor and the instruments is referred to as the Jacobian Equation (JE). The thesis is presented along two main categories: Stable JE and Unstable JE. Under the Stable JE, models with a single and multiple breaks in the Structural Equation (SE) are examined. The break fraction estimators obtained are shown to be consistent for the true break fraction in the model. Additionally, using the fixed break approach, their $T$-convergence rates are established. Monte Carlo simulations which support the asymptotic properties are presented. Two main types of Unstable JE models are considered: a model with a single break only in the JE and another with a break in both the JE and SE. The asymptotic properties of the estimators obtained from these models are intractable under the fixed break approach, hence the thesis provides essential steps towards establishing the properties using the shrinking breaks approach. Nonetheless, a series of Monte Carlo simulations conducted provide strong support for the consistency of the break fraction estimators under the Unstable JE. A combined procedure for testing and estimating significant break points is detailed in the thesis. This method yields a consistent estimator of the true number of breaks in the model, as well as their locations. Lastly, an empirical application of the proposed methodology is presented using the New Keynesian Phillips Curve (NKPC) model for U.S. data. A previous study has found this NKPC model is unstable, having two endogenous regressors with Unstable JE. Using the combined testing and estimation approach, similar break points were estimated at 1975:2 and 1981:1. Therefore, using the GMM estimation approach proposed in this study, the presence of a Stable or Unstable JE does not affect estimations of breaks in the SE. A researcher can focus directly on estimating potential break points in the SE without having to pre-estimate the breaks in the JE, as is currently performed using Two Stage Least Squares.

Bioconjugation of peptides and asymmetric synthesis of gem-difluoromethylene compounds are areas of the modern organic chemistry for which mild and selective methods continue to be developed. This thesis reports new methodologies for these two areas based on the use of stabilized carbenium ions. The reaction that makes the bioconjugation of peptides possible takes place via the direct nucleophilic substitution of alcohols and is driven by the spontaneous formation of stabilized carbenium ions in water. By reacting with the thiol group of cysteine in very mild conditions and with a high selectivity, these carbenium ions allow the site-specific ligation of polypeptides containing cysteine and their covalent derivatization with functionalized probes. The ligation of the indole ring of tryptophan, an emerging target in bioconjugation, is also shown and takes place in the same conditions. The second area investigated is the challenging access to optically active gem-difluoromethylene compounds. We describe a methodology relying on the synthesis of enantioenriched 1,3-benzodithioles intermediates that are shown to be precursors of the corresponding gem-difluoromethylene analogues by oxidative desulfurization-fluorination. This synthesis takes advantage of the highly enantioselective organocatalytic α-alkylation of aldehydes with the benzodithiolylium ion and of the wide possibilities of synthetic transformations offered by the 1,3-benzodithiole group. This approach allows the asymmetric access to complex gem-difluoromethylene compounds through a late-stage fluorination step, thus avoiding the use of fluorinated building blocks.

This thesis describes the synthesis of gem-difluorinated cyclic molecules using building block approaches based mainly on ring-closing metathesis (RCM) using commercially available ruthenium catalysts such as Grubbs' catalyst. In the first instance, 1-bromo-1,1-difluoroprop-2-ene was used to synthesise difluorinated dihydropyrans in order to demonstrate that the unprecedented RCM of a substrate containing two fluorine atoms in the allylic position could be achieved. A similar approach allowed the highly diastereoselective synthesis of new 4,4-difluoro-4-deoxyhexoses using a RCM-dihydroxylation sequence. In order to widen the range of available difluorinated monosaccharide analogues, a potentially highly enantioselective, non-RCM based route was developed. This approach relied on the use of (3-bromo-3,3-difluoro-prop-1-ynyl)-benzene as the fluorinated building block and Sharpless asymmetric dihydroxylations to introduce hydroxyl groups enantioselectively. Unfortunately, a poor choice of protecting group prevented access to the desired difluorinated monosaccharide analogues, even if the asymmetric dihydroxylation proved successful and enantioselective. RCM was also used to synthesise different types of difluorocyclooctenones from trifluoroethanol. These difluorinated 8-membered carbocycles showed interesting and unusual conformational behaviour and were investigated by NMR experiments and a simple computational study. These difluorocyclooctenones were also used to synthesise new bicyclic structures, which are effectively conformationally restrained difluorinated monosaccharide analogues.

Speech intelligibility represents how comprehensible a speech is. It is more important than speech quality in some applications. Single channel speech intelligibility enhancement is much more difficult than multi-channel intelligibility enhancement. It has recently been reported that training-based single channel speech intelligibility enhancement algorithms perform better than Signal to Noise Ratio (SNR) based algorithm. In this thesis, a training-based Deep Neural Network (DNN) is used to improve single channel speech intelligibility. To increase the performance of the DNN, the Multi-Resolution Cochlea Gram (MRCG) feature set is used as the input of the DNN. MATLAB objective test results show that the MRCG-DNN approach is more robust than a Gaussian Mixture Model (GMM) approach. The MRCG-DNN also works better than other DNN training algorithms. Various conditions such as different speakers, different noise conditions and reverberation were tested in the thesis.

A growing body of literature and research emphasizes the importance of developing student mindsets and social and emotional learning (SEL) competencies (metacognitive learning skills) across the prenatal (P) to graduate school (20) (P-20) continuum to increase student motivation, engagement, achievement, and overall well-being. There is, however, an absence of research investigating the impact of braiding growth mindset, SEL, and mindfulness principles and practices together on early elementary student and teacher outcomes. The purpose of my dissertation is to measure the impact of a new PreK–3rd curriculum called Growing Early Mindsets (GEM™) (Coates, in publication) on student and teacher outcomes across the K–3rd continuum in two districts in Oregon. Data collected during the 2014–15 Mindset Works’ study of Growing Early Mindsets (GEM™) was used. Classrooms were assigned to experimental (implemented GEM™) and control groups and given pretest and posttest measures to measure the impact of GEM™ on students’ approaches to learning, social and emotional learning (SEL) competencies, and literacy skills as well as on teacher mindsets, perceptions, confidence, and motivation. Results were statistically significant for students’ approaches to learning and SEL competencies as measured by Teacher Reports and teacher’s beliefs as measured by the Teacher Mindset Survey. Results were not statistically significant for students’ approaches to learning and SEL competencies as measured by Student Surveys, nor students’ oral reading fluency as measured by district-administered oral reading fluency measures. All experimental teachers reported that GEM™ changed their perceptions of their own and others’ learning and growth, increased their confidence to integrate growth mindset and SEL practices, and increased their motivation to improve their overall teaching practices. While the results are somewhat promising, the findings raise many questions that need further exploration.
10000-01-01

Mestrado em Actuarial Science
O pricing na atividade seguradora está a tornar-se cada vez mais interessante e desafi- ador pelo facto de a dimensão dos dados a analisar estar a crescer de forma explosiva. Torna-se assim urgente para as seguradoras reconsiderar a forma de lidar com este vol- ume de dados. Para implementar modelos sofisticados de pricing para produtos de seguro automóvel, aplicámos técnicas de machine learning, incluindo modelos GLM penalizados e métodos de boosting, que ajudam a identificar as características mais importantes de entre uma grande quantidade de variáveis candidatas. Estes métodos também permitem detetar potenciais interações sem testar as inúmeras combinações bidimensionais. Para um uso eficiente desses métodos, é necessário compreender o objetivo do modelo, as hipóteses que o suportam e dominar as metodologias estatísticas. Embora haja alguma evidência de um maior poder preditivo dos modelos baseados em machine learning quando comparados com os tradicionais GLM, estes últimos beneficiam de uma estrutura, mais conveniente e mais interpretável. O modelo GLM é mais fácil de ex- plicar às partes interessadas o que nos levou a utilizar os GLM na modelação do risco, mas absorvendo os ensinamentos dados pelos modelos de machine learning. A avaliação dos modelos é realizada pela análise dos resíduos quer na fase de treino quer de validação quer ainda de teste. Após a revisão pela equipa, aplicam-se alguns ajustes em cada modelo para reforçar a sua significância e a sua robustez. Espera-se que eles tenham alto poder preditivo nos dados fora da amostra e possam, portanto, ser usados no futuro.
Insurance pricing nowadays is getting more and more interesting and challenging due to the fact that the dimension of analysable data is evolutionarily exploding. It is an urgent call for insurers to reconsider how to deal with the data more accurately and precisely. To implement pricing sophistication in motor insurance products, we apply cutting edge machine learning techniques including penalized GLM and boosting methods, which help us identify the important features among massive amount of candidate variables, and detect potential interactions without trying the endless two-way combinations manually. In order to sufficiently make use of these methods, we need to deeply understand the research objective, preliminary assumptions and statistical backgrounds. Although there is some evidence indicating the existence of higher predictive power of machine learning models compared with traditional GLM (Generalized Linear Models), GLM is more convenient and interpretable, especially for multiplicative models. GLM model is easier to be demonstrated to stakeholder, therefore we still achieve our risk models in GLM, but absorbing the insights from our machine learning results. The evaluation of models is done by progression, it is generally performed by residual analysis of the training or validation dataset, and testing errors for the holdout dataset. After peer review, we apply some adjustment in each model, to get models that are significant and robust. They are expected to have high predictive power in the out-of- sample data, thus can be used in the future.
info:eu-repo/semantics/publishedVersion

Availability of precipitation data is very important in every aspect related to hydrology. Readings from the ground stations are reliable and are used in hydrological models to do various analysis. However, the predictions are always associated with uncertainties due to the limited number of ground stations, which requires interpolation of the data. Meanwhile, groundbreaking approach in capturing precipitation events from vantage point through satellites in space has created a platform to not only merge ground data with satellite estimates to produce more accurate result, but also to find the data where ground stations are not available or scarcely available. Nevertheless, the data obtained through these satellite missions needs to be verified on its temporal and spatial resolution as well as the uncertainties associated before we make any decisions on its basis. This study focuses on finding and evaluating data obtained from two multi-satellite precipitation measurements missions: i) Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA) ii) Global Precipitation Measurement (GPM) mission. GPM is the latest mission launched on Feb 28, 2014 after the successful completion of TRMM mission which collected valuable data for 17 years since its launch in November 1997. Both near real time and final version precipitation products for TMPA and GPM are considered for this study. Two study areas representing eastern and western parts of the United States of America (USA) are considered: i) Charlotte (CLT) in North Carolina ii) San Francisco (SF) in California. Evaluation is carried out for daily accumulated rainfall estimates and single rainfall events. Statistical analysis and error categorization of daily accumulated rainfall estimates were analyzed in two parts: i) Ten yeas data available for TMPA products were considered for historical analysis ii) Both TMPA and GPM data available for a ten-month common period was considered for GPM Era analysis. To study how well the satellite estimates with their finest temporal and spatial resolution capture single rainfall event and to explore their engineering application potential, an existing model of SF watershed prepared in Infoworks Integrated Catchment Model (ICM) was considered for hydrological simulation. Infoworks ICM is developed and maintained by Wallingford Software in the UK and SF watershed model is owned by San Francisco Public Works (SFPW). The historical analysis of TMPA products suggested overestimation of rainfall in CLT region while underestimation in SF region. This underestimation was largely associated with missed-rainfall events and negative hit events in SF. This inconsistency in estimation was evident in GPM products as well. However, in the study of single rainfall events with higher magnitude of rainfall depth in SF, the total rainfall volume and runoff volume generated in the watershed were over-estimated. Hence, satellite estimates in general tends to miss rainfall events of lower magnitude and over-estimate rainfall events of higher magnitude. From statistical analysis of GPM Era data, it was evident that GPM has been able to correct this inconsistency to some extent where it minimized overestimation in CLT region and minimized negative error due to underestimation in SF. GPM products fairly captured the hydrograph shape of outflow in SF watershed in comparison to TMPA. From this study, it can be concluded that even though GPM precipitation estimates could not quiet completely replace ground rain gage measurements as of now, with the perpetual updating of algorithms to correct its associated error, it holds realistic engineering application potential in the near future.

The main aim of this thesis is to explain the inflation dynamics in Turkey within a theoretically consistent empirical framework. The New Keynesian Phillips Curve (NKPC) is chosen as the basis model for our analysis because, by describing the inflation process within an intertemporal optimizing dynamic general equilibrium model, it provides a rigorous analytical groundwork for credible welfare and policy analysis. We have contributed to the literature by developing a NKPC formulation that is novel in the literature: A constant elasticity of substitution (CES) type of production function incorporating imported and domestically produced intermediate goods was combined with incomplete exchange rate pass through to import prices. The short-run inflation dynamics were analyzed within the context of this new specification by estimating the model&rsquo
s highly nonlinear structural parameters that capture the price-setting behavior in Turkey for period 1988:1 - 2009:4. Our findings suggest that this NKPC formulation can explain the 1994 and 2000-01 crises as well as the current environment of low inflation achieved with the adoption of the implicit and fully fledged inflation targeting regimes quite well. As a policy application we explored the effects of the inflation targeting framework adopted after the 2000-01 crises on the parameters characterizing the inflation process in Turkey. The subsample econometric results suggested that the inflation targeting framework applied was quite successful in decreasing inflation inertia in Turkey. Thus, should the success of the inflation targeting regime continue, this should be taken as an opportunity to reduce inflation substantially with very low output losses.

En los últimos años, las mujeres están siendo reconocidas como un factor importante en la creación de nuevas empresas y en la promoción de desarrollo económico. Aunque, a pesar de los esfuerzos realizados desde diferentes ámbitos, las cifras de emprendimiento femenino siguen siendo menores a la de sus homólogos masculinos. El principal objetivo de esta investigación es analizar la influencia de los factores del entorno en la actividad emprendedora femenina en Cataluña (España). Para conseguir este fin, se usa la Teoría Económica Institucional (North, 1990 and 2005) como marco conceptual. Explícitamente, este artículo: 1) explora, identifica y analiza cualitativamente los factores del entorno (instituciones formales e informales) que influyen en la actividad emprendedora femenina en Cataluña, 2) explora y lleva a cabo un análisis cualitativo de los factores socio-culturales (instituciones informales) que influyen en el emprendimiento femenino en Cataluña; 3) analiza cuantitativamente los factores del entorno (instituciones formales e informales) que influyen en la actividad emprendedora femenina en España en el año 2010, 4) analiza cuantitativamente los factores socio-culturales (instituciones informales) que influyen en la actividad emprendedora femenina en los años 2009 y 2010; y 5) analiza cuantitativamente los factores del entorno (instituciones formales e informales) que influyen en el emprendimiento femenino en España a nivel regional en el período 2003-2010. Las metodologías usadas para dar cumplimiento a los objetivos incluyen tanto una metodología cualitativa que nos ha de permitir explorar y obtener información que, a su vez, nos permita comprender la naturaleza del comportamiento de las mujeres emprendedoras, así como los factores que influyen en dicho comportamiento. Por otra parte, también se incluyen metodologías cuantitativas que nos van a permitir obtener una mayor generalización y objetivación de los resultados a través de las muestras utilizadas y hacer inferencia al total de mujeres emprendedoras de Cataluña. Esta investigación proporciona un método para analizar los factores del entorno (factores institucionales) que influyen en la actividad emprendedora femenina, a nivel individual, regional y nacional, combinando muestras de Global Entrepreneurship Monitor (GEM), Adult Population Survey (APS) y National Expert Survey (NES) con información del Instituto Nacional de Estadística (INE). Los principales resultados de la investigación revelan que, aunque tanto los factores formales como informales afectan al emprendimiento, el efecto de las instituciones informales (redes sociales, modelos de referencia, miedo al fracaso, percepción de las habilidades, carrera emprendedora y contexto familiar), son más importantes que los efectos de las instituciones formales (financiación, políticas y ayudas no económicas, diferencias salariales basadas en el género, tamaño familiar o nivel educativo) en el caso de las mujeres emprendedoras de Cataluña y España. Esta investigación sugiere una serie de implicaciones a nivel académico, al posicionar la economía institucional como un marco conceptual apropiado para el análisis de los factores del entorno que contribuyen a la promoción de la actividad emprendedora femenina en Cataluña y España. Desde una perspectiva práctica, este estudio puede ayudar a generar políticas que mejoren la percepción de las habilidades emprendedoras, incrementar la visibilidad de los modelos de referencia femeninos, y favorecer la integración del contexto familiar de las mujeres emprendedoras
In recent years, women have begun to be recognized as an important factor in the creation of new businesses and in the promotion of economic development. However, despite the efforts made in various fields, the number of female entrepreneurs continues to be lower than that of their male counterparts. The main objective of this investigation is to analyse the influence of environmental factors on female entrepreneurial activity in Catalonia (Spain). To achieve this end, institutional economic theory (North, 1990, 2005) is used as the theoretical framework. Thus, this study explicitly: 1) explores, identifies and qualitatively analyses the environmental factors (formal and informal institutions) that influence female entrepreneurship in Catalonia; 2) explores and conducts a qualitative analysis of the socio-cultural factors (informal institutions) that influence female entrepreneurship in Catalonia; 3) quantitatively analyses the environmental factors (formal and informal institutions) that influenced female entrepreneurship in Spain in the year 2010; 4) quantitatively analyses the socio-cultural factors (informal institutions) that influenced female entrepreneurship in Catalonia in the years 2009 and 2010; and 5) quantitatively analyses the environmental factors (formal and informal institutions) that influenced female entrepreneurship in Spain at a regional level for the period 2003–2010. The methodologies used in order to achieve the objectives include a qualitative methodology that should permit us to obtain information that, in turn, will allow us to understand better the nature of the behaviour of female entrepreneurs and the factors that influence their behaviour. Furthermore, the methodologies used include a quantitative method that will allow us to obtain greater generalization and objectification of the results by way of the samples used, which should permit us to make inferences regarding female entrepreneurs in Catalonia. This investigation provides a method for analysing the environmental factors (institutional factors) that influence female entrepreneurial activity at an individual, regional and national level, combining data from the Global Entrepreneurship Monitor (GEM) and from both the Adult Population Survey (APS) and the National Expert Survey (NES) with information from the National Statistics Institute (INE). The main findings of the research reveal that, while both formal and informal factors affect entrepreneurship, the effect of informal institutions (social networks, role models, fear of failure, perceived capabilities, entrepreneurial career and family context) is more significant than the effect of formal institutions (financing, non-economic policies and assistance, gender-based pay gap, family size or education) in the case of female entrepreneurs in Catalonia and Spain. This investigation suggests a series of implications at the academic level, as it positions institutional economics as an appropriate conceptual framework for the analysis of the environmental factors that contribute to the promotion of female entrepreneurial activity in Catalonia and Spain. From the practical perspective, this study may help to generate policies that improve the perception of entrepreneurial capabilities, increase the visibility of female role models and favour the integration of the female entrepreneurs’ family context.

Due to the rapid development and growing need for information technologies, more and more researchers start to focus on high-dimensional data. Much work has been done on problems like point estimation possessing oracle inequalities, coefficient estimation, variable selection in high-dimensional regression models. However, with respect to the statistical inference for the regression coefficients, there have been few studies. Therefore, we propose a regularized efficient score estimation and testing (RESET) approach for treatment effects in the presence of nuisance parameters, either low-dimensional or high-dimensional, in generalized linear models (GLMs). Based on the RESET method, we are also able to develop another two-step approach related to the same problem. The RESET approach is based on estimating the efficient score function of the treatment parameters. This means we are trying to remove the influence of nuisance parameters on the treatment parameters and construct an efficient score function which could be used for estimating and testing for the treatment effect. The RESET approach can be used in both low-dimensional and high-dimensional settings. As the simulation results show, it is comparable with the commonly used maximum likelihood estimators in most low-dimensional cases. We will prove that the RESET estimator is consistent under some regularity conditions, either in the low-dimensional or the high-dimensional linear models. Also, it is shown that the efficient score function of the treatment parameters follows a chi-square distribution, based on which the regularized efficient score tests are constructed to test for the treatment effect, in both low-dimensional and high-dimensional GLMs. The two-step approach is mainly used for high-dimensional inference. It combines the RESET approach with a first step of selecting "promising" variables for the purpose of reducing the dimension of the regression model. The minimax concave penalty is adopted for its oracle property, which means it tends to choose "correct" variables asymptotically. The simulation results show that some improvement is still required for this approach, which will be part of our future research direction. Finally, both the RESET and the two-step approaches are implemented with a real data example to demonstrate their application, followed by a conclusion for all the problems investigated here and a discussion for the directions of future research.

In questa tesi viene ripercorsa la storia e l'evoluzione della teoria degli stakeholder, della responsabilità sociale di impresa e del bilancio sociale. Focalizzando l'attenzione su quest'ultimo, è stato riportato il bilancio sociale provvisorio di Conapi Soc. Coop. Agricola, ideato ed elaborato con la cooperativa e redatto secondo lo standard nazionale GBS.

The acidification of soils has been a continuous process since at least the beginning of the 20th century. Therefore, an inquiry of how and when the soil pH levels have changed is relevant to gain better understanding of this process. The aim of this thesis is to study the average national soil pH level over time in Sweden and the local spatial differences within Sweden over time. With data from the Swedish National Forest Inventory, soil pH surfaces are estimated for each surveyed year together with the national average soil pH using a generalized additive modeling approach with one model for each pair of consecutive years. A decreasing trend in average national level soil pH was found together with some very weak evidence of year-to-year differences in the spatial structure of soil pH.

Measurement processes are vital for any organization as they are used to asses, analyze, monitor and control the processes, products and resources. The measurement programs are used in different ways in different organizations. Most of the measurement programs fail to provide the expected results; therefore it‘s needed to consider the success factors and reasons of failures for the measurement programs. The GQM is the most widely used model for measurement programs having various extensions to overcome its shortcomings in different scenarios. The Goals, Questions and metrics are defined and used in different ways at different levels in the organizations. There is a need of improving the measurement programs and one of the solutions is to provide a framework that can define the goals, questions and measures in a structured way. The prioritization, traceability and re-usability of goals and questions provide the effectiveness in the measurement program. The optimization of the measures and building a measurement repository makes the measurement collection process precise. In this thesis, the results of a systematic review on the current literature on software measurement programs are presented. An assessment of the current state of art on measurement programs, their usability and success factors is done. The study of measurement models, frameworks, tools and standards is done later on to know the different ways of goals, questions definition and measurement collection methods. The systematic review of the research work is done over the period 1997 – 2009. In order to understand and explore the difficulties in application of measurement programs in the industrial settings, interviews are conducted within a CMMI Level 3 company. On the basis of the systematic review analysis results and industrial interviews, a framework for a more effective measurement process is defined and within the framework, a model called ‗Structured Prioritized Goal Question Metrics (SPGQM)‘ is developed. This framework extends the well-known Goal Question Metric paradigm and basically comprises of two models; the Optimum Measures Set Decision (OMSD) model developed within a Master of Science thesis study at the Blekinge Institute of Technology and the SPGQM. This framework defines the process in order to define structured goals and questions with the help of templates and to prioritize them with the help of OMSD model. This framework has been validated in a CMMI Level 3 company. The validation was done by means of conducting a case study.

Dans le cadre de cette thèse, notre préoccupation principale est de développer des approches non supervisées permettant de résoudre des problèmes de grande taille le plus efficacement possible. Pour ce faire, nous avons considéré des approches bayésiennes qui permettent d'estimer conjointement les paramètres de la méthode avec l'objet d'intérêt. Dans ce cadre, la difficulté principale est que la loi a posteriori est en général complexe. Pour résoudre ce problème, nous nous sommes intéressés à l'approximation bayésienne variationnelle (BV) qui offre une approximation séparable de la loi a posteriori. Néanmoins, les méthodes d’approximation BV classiques souffrent d’une vitesse de convergence faible. La première contribution de cette thèse consiste à transposer les méthodes d'optimisation par sous-espace dans l'espace fonctionnel impliqué dans le cadre BV, ce qui nous permet de proposer une nouvelle méthode d'approximation BV. Nous avons montré l’efficacité de notre nouvelle méthode par les comparaisons avec les approches de l’état de l’art.Nous avons voulu ensuite confronter notre nouvelle méthodologie à des problèmes de traitement d'images de grande taille. De plus nous avons voulu favoriser les images régulières par morceau. Nous avons donc considéré un a priori de Variation Total (TV) et un autre a priori à variables cachées ressemblant à un mélange scalaire de gaussiennes par changement de positions. Avec ces deux modèles a priori, en appliquant notre méthode d’approximation BV, nous avons développé deux approches non-supervisées rapides et bien adaptées aux images régulières par morceau.En effet, les deux lois a priori introduites précédemment sont corrélées ce qui rend l'estimation des paramètres de méthode très compliquée : nous sommes souvent confronté à une fonction de partition non explicite. Pour contourner ce problème, nous avons considéré ensuite de travailler dans le domaine des ondelettes. Comme les coefficients d'ondelettes des images naturelles sont généralement parcimonieux, nous avons considéré des lois de la famille de mélange scalaire de gaussiennes par changement d'échelle (GSM) pour décrire la parcimonie. Une autre contribution est donc de développer une approche non-supervisée pour les lois de la famille GSM dont la densité est explicitement connue, en utilisant la méthode d'approximation BV proposée
In this thesis, our main objective is to develop efficient unsupervised approaches for large dimensional problems. To do this, we consider Bayesian approaches, which allow us to jointly estimate regularization parameters and the object of interest. In this context, the main difficulty is that the posterior distribution is generally complex. To tackle this problem, we consider variational Bayesian (VB) approximation, which provides a separable approximation of the posterior distribution. Nevertheless, classical VB methods suffer from slow convergence speed. The first contribution of this thesis is to transpose the subspace optimization methods to the functional space involved in VB framework, which allows us to propose a new VB approximation method. We have shown the efficiency of the proposed method by comparisons with the state of the art approaches. Then we consider the application of our new methodology to large dimensional problems in image processing. Moreover, we are interested in piecewise smooth images. As a result, we have considered a Total Variation (TV) prior and a Gaussian location mixture-like hidden variable model. With these two priors, using our VB approximation method, we have developed two fast unsupervised approaches well adapted to piecewise smooth images.In fact, the priors introduced above are correlated which makes the estimation of regularization parameters very complicated: we often have a non-explicit partition function. To sidestep this problem, we have considered working in the wavelet domain. As the wavelet coefficients of natural images are generally sparse, we considered prior distributions of the Gaussian scale mixture family to enforce sparsity. Another contribution is therefore the development of an unsupervised approach for a prior distribution of the GSM family whose density is explicitly known, using the proposed VB approximation method

Une carte génétique densément peuplée en marqueurs SNP a été construite pour le mil en utilisant une population de 93 individus F2. D’abord, un nouveau protocole de génotypage par séquençage (GBS) impliquant une combinaison d’enzymes de restriction PstI-MspI avec des amorces PCR incluant trois bases sélectives a été utilisé pour obtenir 3 321 marqueurs SNP. Suite à plusieurs procédés de filtration, 314 marqueurs SNP non-redondants distribués sur sept groupes de liaison ont permis de construire une carte couvrant une distance génétique totale de 640 cM avec un intervalle moyen de 2,1 cM entre marqueurs SNP. La taille des groupes de liaison variait considérablement (62 cM à 123 cM). Finalement, 19 marqueurs SSR ont été analysés sur les parents de la population afin de trouver ceux qui étaient polymorphes et ainsi faire le pont entre notre carte et celles rapportées précédemment. Parmi eux, quatre marqueurs SSR étaient polymorphes et ont permis d’établir une correspondance entre quatre groupes de liaison de notre carte et ceux des cartes antérieures. La disponibilité d’une telle carte peut être un atout exploitable pour l’identification de régions associées à certains caractères d’importance agronomique.
A high-density SNP genetic map was constructed for pearl millet using a mapping population of 93 F2 individuals. Firstly, a Genotyping by Sequencing (GBS) protocol involving a combination of restriction enzymes PstI-MspI with PCR primers including three selective bases was used to obtain 3,321 SNP markers. Following several filtration processes, 314 non-redundant SNPs distributed over seven linkage groups were used to build a map covering a total genetic distance of 640 cM with an average interval of 2.1 cM between SNPs. The size of the linkage groups varied considerably (62 cM to 123 cM). Finally, 19 SSR markers were analyzed on the parents of the population in order to find those that were polymorphic and thus bridge the gap between our map and those reported previously. Among them, four SSR markers were polymorphic and have established a correspondence between four linkage groups in our map and previous maps. The availability of such a map may be exploitable for the identification of regions associated to some important agronomic characters.

To support video communication over mobile environments has been one of the objectives of many engineers of telecommunication networks and it has become a basic requirement of a third generation of mobile communication systems. This dissertation explores the possibility of optimizing the utilization of shared scarce radio channels for live video transmission over a GSM (Global System for Mobile telecommunications) network and realizing error resilient video communication in unfavorable channel conditions, especially in mobile radio channels. The main contribution describes the adoption of a SEC (Second Error Correction) approach using ECC (Error Correction Coding) based on a Punctured Convolutional Coding scheme, to cope with residual errors at the application layer and enhance the error resilience of a compressed video bitstream. The approach is developed further for improved performance in different circumstances, with some additional enhancements involving Intra Frame Relay and Interleaving, and the combination of the approach with Packetization. Simulation results of applying the various techniques to test video sequences Akiyo and Salesman are presented and analyzed for performance comparisons with conventional video coding standard. The proposed approach shows consistent improvements under these conditions. For instance, to cope with random residual errors, the simulation results show that when the residual BER (Bit Error Rate) reaches 10-4, the video output reconstructed from a video bitstream protected using the standard resynchronization approach is of unacceptable quality, while the proposed scheme can deliver a video output which is absolutely error free in a more efficient way. When the residual BER reaches 10-3, the standard approach fails to deliver a recognizable video output, while the SEC scheme can still correct all the residual errors with modest bit rate increase. In bursty residual error conditions, the proposed scheme also outperforms the resynchronization approach. Future works to extend the scope and applicability of the research are suggested in the last chapter of the thesis.

To support video communication over mobile environments has been one of the objectives of many engineers of telecommunication networks and it has become a basic requirement of a third generation of mobile communication systems. This dissertation explores the possibility of optimizing the utilization of shared scarce radio channels for live video transmission over a GSM (Global System for Mobile telecommunications) network and realizing error resilient video communication in unfavorable channel conditions, especially in mobile radio channels. The main contribution describes the adoption of a SEC (Second Error Correction) approach using ECC (Error Correction Coding) based on a Punctured Convolutional Coding scheme, to cope with residual errors at the application layer and enhance the error resilience of a compressed video bitstream. The approach is developed further for improved performance in different circumstances, with some additional enhancements involving Intra Frame Relay and Interleaving, and the combination of the approach with Packetization. Simulation results of applying the various techniques to test video sequences Akiyo and Salesman are presented and analyzed for performance comparisons with conventional video coding standard. The proposed approach shows consistent improvements under these conditions. For instance, to cope with random residual errors, the simulation results show that when the residual BER (Bit Error Rate) reaches 10-4, the video output reconstructed from a video bitstream protected using the standard resynchronization approach is of unacceptable quality, while the proposed scheme can deliver a video output which is absolutely error free in a more efficient way. When the residual BER reaches 10-3, the standard approach fails to deliver a recognizable video output, while the SEC scheme can still correct all the residual errors with modest bit rate increase. In bursty residual error conditions, the proposed scheme also outperforms the resynchronization approach. Future works to extend the scope and applicability of the research are suggested in the last chapter of the thesis.

La toxine bactérienne de Shiga se lie au glycosphingolipide (GSL) globotriaosylcéramide (Gb3) afin d’entrer par endocytose dans les cellules en utilisant une voie dépendante et indépendante de la clathrine. Dans la voie indépendante de la clathrine, la toxine de Shiga réorganise les lipides de la membrane de façon à imposer une contrainte mécanique sur la bicouche, conduisant ainsi à la formation de pic d’invagination d'endocytose profonds et étroits. Mécaniquement ce phénomène n’est pas encore compris, notamment il reste énigmatique, comment se traduisent les propriétés géométriques de l’agrégation des glycosphingolipides GSLS et de la toxine. Dans mon travail de thèse, via l’utilisation de la sous-unité B de la toxine de Shiga (STxB) comme un modèle, différentes espèces moléculaires de son récepteur Gb3 ont été synthétisés avec des structures délibérément choisis. Les études réalisées par imagerie de haute résolution et par la modélisation informatique ont permis d’élucider les contraintes mécano-chimique sous-jacente conduisant à une réorganisation efficace qui a pour résultat l’agrégation de la toxine et la réorganisation des lipides. En combinant des expériences de simulation sur ordinateur de dynamique des particules dissipatives (DPD) et des expériences sur des modèles de membranes cellulaires, nous avons fourni la preuve de l’induction d’une force de fluctuation-membrane, de type « force de Casimir », conduisant à l'agrégation des molécules de toxines associées à la membrane à des échelles de longueur mésoscoiques. Nous avons observé et mesuré, en outre la condensation lipidique induite par la toxine, quantitativement sur des monocouches de Langmuir en utilisant la réflectivité des rayons X (XR) et par la mesure de la diffraction des rayons X par incidence rasante (GIXD), fournissant ainsi une preuve directe de l'hypothèse que la toxine a le potentiel de réduire de façon asymétrique la surface moléculaire sur la partie membranaire exoplasmique, ce qui conduit à une déformation locale de la membrane. Durant ma thèse, nos efforts ont été consacrés à la réalisation de nouveaux glycosphinolipides (GSL) comme outils chimiques à visée biologique. Par ailleurs, une nouvelle stratégie de reconstitution de GSL fonctionnels sur la membrane cellulaire, basée sur une réaction de ligation de type « click » entre un glycosyl-cyclooctyne et un azido-sphingosine a été étudiée. Les résultats obtenus sur les cellules se sont avérés beaucoup moins efficace que ceux in vitro. Une poursuite de l'optimisation de cette méthodologie est actuellement en cours. Une sonde fluorescente du glycosphinolipide Gb3, marquée à l’Alexa Fluor 568 lui-même lié par l'intermédiaire d'un bras PEG-α à la position de la chaîne acyle, a été synthétisée. Cette sonde se lie à la STxB sur couche mince de TLC, mais pas sur des membranes modèles. D'autres améliorations sont discutées
Bacterial Shiga toxins bind to the glycosphingolipid (GSL) globotriaosylceramide (Gb3) to enter cells by clathrin-dependent and independent endocytosis. In the clathrin-independent pathway, Shiga toxin reorganizes membrane lipids in a way such as to impose mechanical strain onto the bilayer, thus leading to the formation of deep and narrow endocytic pits. Mechanistically how this occurs is not yet understood, and notably how the geometric properties of toxin-GSLs complexes translate into function has remained enigmatic. In my thesis work, using the B-subunit of Shiga toxin (STxB) as a model, different molecular species of its receptor Gb3 have been synthesized with deliberately chosen structures, coupled with high resolution imaging and computational modeling, to understand the underlying mechano-chemical constraints leading to efficient toxin clustering and lipids reorganization. By combining dissipative particle dynamics (DPD) computer simulation and experiments on cell and model membranes, we provided evidence that a membrane fluctuation-induced force, termed Casimir-like force, drives the aggregation of tightly membrane-associated toxin molecules at mesoscopic length scales. Furthermore, toxin-induced lipid condensation was observed and measured quantitatively on Langmuir monolayers using X-ray reflectivity (XR) and grazing incidence x-ray diffraction (GIXD), thereby providing direct evidence for the hypothesis that the toxin has the potential to asymmetrically reduce the molecular area of the exoplasmic membrane leaflet, leading to local membrane deformation. During my PhD, effort was also invested to develop new GSL tools applied to the biological setting. A novel strategy based on the Cu-free click reaction between glycosyl-cyclooctyne and azido-sphingosine was designed with the goal to functionally incorporate GSLs into cellular membranes. Following the synthesis work, click reactions have been performed in solution and on cells. Compared to the former, results on cells were far less efficient. Further optimization is currently ongoing. A fluorescently labeled Gb3 probe with Alexa Fluor 568 coupled via a PEG linker to the α-position of the acyl chain, was synthesized, to which STxB bound on TLCs, but not on model membranes. Further improvements are discussed

Les polymères en solution peuvent s’auto-assembler pour former différents types de complexes (coacervats, agrégats, etc.). Ces objets, en plus de régir l’organisation cellulaire peuvent être détournés pour encapsuler des molécules d’intérêt. Cependant, avant de pouvoir les utiliser en tant que tel, la compréhension des mécanismes d’auto-assemblage est indispensable pour pouvoir les contrôler et ainsi former des objets à façon. Ce travail de thèse porte sur l’auto-assemblage entre la gomme d’Acacia et le chitosane. Les interactions mises en jeu lors de la séparation de phases associatives reposent principalement sur des interactions électrostatiques. Dans un premier temps, la formation de ce complexe a été analysée d’un point de vue thermodynamique par titration calorimétrique isotherme, en fonction des paramètres physico-chimiques, en particulier du pH, de la température et du ratio molaire. Il a ainsi été démontré qu’une augmentation de température ou de pH favorisait l’enthalpie au dépend de l’entropie et qu’il était possible de former des coacervats dans toutes les conditions étudiées en jouant sur le couple pH/température. Pour changer le nombre et la nature des interactions entre les polysaccharides, ces derniers ont été modifiés par voie enzymatique. Un procédé développé au laboratoire reposant sur l’utilisation d’une laccase pour oxyder l’acide férulique a été mis en œuvre. Dans un premier temps, la modification de la gomme d’Acacia a été démontrée, notamment grâce à des outils spectroscopiques (FTIR, RMN). De plus, la structure des nombreux produits d’oxydation a été étudiée (RMN, LCMS). Les produits obtenus sont majoritairement des dimères d’acide férulique et une structure complète a été élucidée. Les produits greffés sur la gomme étant de nature phénolique, leur greffage a un impact sur les propriétés techno-fonctionnelles de cette dernière (solubilité dans l’eau, comportements thermique et rhéologique, tension de surface, propriétés antioxydantes, etc.). Enfin, le greffage de composés phénoliques modifiant l’hydrophobie et les charges des polysaccharides, des auto-assemblages entre les différents polymères ont été réalisés pour déterminer leur impact sur les interactions mises en jeu et la nature des complexes formés. Dans des conditions données et en fonction des polysaccharides mis en présence, il est possible de former des coacervats, des agrégats voire un mélange des deux. La formation de ces objets a été corrélée avec une étude des paramètres thermodynamiques mis en jeu. Ces résultats ouvrent de nouvelles voies dans la compréhension et le contrôle de la complexation entre polysaccharides qui, à plus long terme, déboucheront sur des applications dans le domaine de l’encapsulation
Polymers in solution can self-assemble to form different types of complexes (coacervates, aggregates, etc.). These objects, in addition to being part of the cellular organization, can be repurposed to encapsulate molecules of interest. However, before being able to use them as such, the understanding of the self-assembly mechanisms is essential to be able to control them and thus form targeted objects. This thesis work focuses on the self-assembly between gum Arabic and chitosan. The interactions involved in associative phase separation are mainly based on electrostatic interactions. Initially, the formation of this complex was analyzed from a thermodynamic point of view by isothermal calorimetric titration, depending on the physicochemical parameters, in particular the pH, the temperature and the molar ratio. It was thus demonstrated that an increase in temperature or pH favored enthalpy and disfavored entropy and that it was possible to form coacervates under all studied conditions by controlling the pH/temperature combination. To change the number and nature of the interactions between polysaccharides, they were modified by enzymatic pathway. A process developed at the laboratory, based on the use of a laccase to oxidize ferulic acid, has been implemented. First, the modification of gum Arabic was demonstrated, in particular with the support of spectroscopic tools (IR-FT, NMR). In addition, the structure of the numerous oxidation products has been studied (NMR, LCMS). The products obtained are mainly ferulic acid dimers and a complete structure has been elucidated. The products grafted on the gum are of phenolic nature so their grafting has an impact on its techno-functional properties (water solubility, thermal and rheological behavior, surface tension, antioxidant properties, etc.). Finally, the grafting of phenolic compounds modified the polysaccharides hydrophobia and charges, so they were mixed in order to determine the impact of these new groups on the interactions involved in the self-assemblies and the nature of the complexes formed. Under given conditions and depending on the polysaccharides used, it is possible to form coacervates, aggregates or even a mixture of both. The formation of these objects was correlated with a study of the thermodynamic parameters involved. These results paves the way for understanding and controlling complexation between polysaccharides that, in the longer term, will lead to applications in the field of encapsulation

La Leustroducsine B, antibiotique apparenté aux Phoslactomycines, intervient dans la régulation des cytokines et constitue avant tout un outil important en biochimie. A ce jour il existe deux synthèses totales et deux synthèses formelles de cette molécule. L'étude de sa synthèse totale représente pour nous l'opportunité d'explorer la chimie des oxazines-l,2. En effet, notre approche se distingue des précédentes par le plan rétrosynthétique convergent que nous proposons via une oxazine chirale fonctionnalisée comme synthon clef qui proviendrait d'une réaction de nitroso Diels-Alder asymétrique et régiosélective. Les diènes-1,3 substitués en position 1,2, précurseurs de ces oxazines, ont été préparés selon des séquences que nous avons mises en œuvre. La cycloaddition de ceux-ci, en séries racémique et chirale, avec respectivement le Boc-nitroso et l'α-chloronitroso chiral dérivé du D-xylose de Wightman, a fourni l'oxazine centrale diversement fonctionnalisée. Les facteurs dirigeant la régiosélectivité de ces réactions ont été étudiés par l'analyse des modèles moléculaires et de la modélisation des états de transition. Cet outil a aussi permis d'interpréter l'origine des bonnes énantiosélectivités obtenues avec cet α-chloronitroso chiral, dont ce travail décrit les premières applications à des diènes acycliques. Cette méthodologie a permis la synthèse de l'oxazine centrale de configuration requise, de façon régiospécifique et avec une bonne énantiosélectivité (10/1 à 15/1). La rupture de la liaison N-O d'oxazines-1,2 (dihydro-3,6 ou tetrahydro) dans des conditions douces, qui constitue l'autre étape clé de notre stratégie de synthèse de la Leustroducsine B, a été étudiée sous deux aspects. D'abord, nous avons voulu poursuivre le développement de la réaction avec des dérivés carbonylés substitués par un hétéroatome en α en utilisant le diméthoxyacétaldéhyde en milieu aqueux. Mais celui-ci, s'il permet le clivage de la liaison N-O dans des conditions neutres, n'a cependant conduit qu'à la formation surprenante d'hémiaminals et d'amides, réaction aussi réalisable avec des aldéhydes simples. Par ailleurs, nous avons observé une réaction tandem cycloaddition-rupture de la liaison N-O, initiée par l'α-chloronitroso chiral de Wightman, que nous avons étudiée sur divers exemples et qui ouvre des perspectives intéressantes pour la synthèse asymétrique des amino alcools correspondant aux dihydro-3,6-oxazines-1,2 intermédiaires du fait qu'elle permet de réaliser ce clivage en présence de fonctions incompatibles avec d'autres méthodologies développées antérieurement. Lors de nos essais de rupture de la liaison N-O d'oxazines-1,2 par des aldéhydes, nous avons mis en évidence des réactions d'évolution de ces composés initiées par l'oxygène dissous dans l'eau. Les structures des principaux produits de ces réactions surprenantes ont été déterminées
Leustroducsin B, an antibiotic related to Phoslactomycins, regulates cytokins production and is an important new tool in biochemistry. At present, only two total and two formaI syntheses have been achieved. Present studies related to its total synthesis had for aim also to explore some chemistry of 1,2-oxazines considered as a protective group of the required amino alcohol of Leustroducsin B. The present synthetic approach differs from the previous ones since, in a convergent scheme, our retrosynthetic analysis involves a functionalized chiral 3,6-dihydro-1,2-oxazine as a key central unit which might be obtained by a regioselective asymmetric nitroso Diels-Alder reaction. The first syntheses of the functionalized (Z)-1,3-dienes-1,2-substituted, precursors of the desired dihydrooxazines, have been achieved. Their cycloaddition, in racemic and chiral series, respectively with BocNO and Wightman's chiral α-chloronitroso derived from D-xylose, afforded diverse functionalized oxazines targeted as possible key-compounds for further couplings with other subunits in a convergent approach. Factors controlling the regioselectivity of the cycloadditions with BocNO and Wightman's chiral α-chloronitroso, are discussed and steric factors analysed with molecular modelisation of the different possible approaches for the cycloaddition. The present work also describes the first examples of cycloadditions of that chiral α-chloronitroso with diverse acyclic dienes, and the regioselectity and enantioselectivity of these reactions are discussed for steric effects with molecular modelisation. The present methodology allowed the synthesis of the targeted chiral 1,2-oxazines, with the required absolute configuration, which are obtained in good yield in regiospecific and highly enantioselective (er = 10 : 1 to 15 : 1) reactions. The N-O bond cleavage in mild conditions compatible with functionalized 1,2-oxazines (3,6-dihydro or tetrahydro) was also an other aim in the present synthetic approach to Leustroducsin B. Ln a first approach, in an attempt to extend and simplify a previous N-O bond cleavage methodology developed in our group by reaction of 1,2-oxazines with carbonyl compounds α-substituted by an heteroatom, a study was done by using the commercially available aqueous solution of dimethoxyacetaldehyde. However, if this aldehyde allows the cleavage of the N-O bond in mild neutral conditions, quite surprisingly only stable hemiaminals and amides are the only products and the free amino alcohol is not observed. Ln the same conditions, reactions with simple aldehydes (isobutyraldehyde or pivaldehyde) only afford the corresponding hydroxy amides. On the other hand, in the present study with Wightman's α-chloronitroso, a tandem reaction nitroso Diels-Alder/N-O bond cleavage was observed at room temperature in anhydrous conditions and further developed on different examples, thus allowing an interesting direct access for the asymmetric synthesis of amino alcohols corresponding to the intermediate 3,6-dihydro-1 ,2-oxazines with no isomerization of the double bond and no loss of enantioselcctivity. Moreover, it is shown that this tandem reaction allows achieving the N-O bond cleavage in the presence of functions which are incompatible with previous methodologies. Reactions of 1,2-oxazines with dissolved oxygen in aqueous solutions, or in the air were also observed and the structures of the products determined for those previously non reported reactions

En esta tesis se han desarrollado los siguientes métodos numéricos para el análisis de estructuras planares, especialmente finline, en la banda de microondas y ondas milimétricas:
-el método de las líneas con discretizacion uniforme y no uniforme.
-el método de los elementos finitos aplicado al análisis de guías homogéneas e inhomogeneas.
-el método de la matriz de lineas de transmisión tridimensional.
-el método de la matriz se generalizada para el análisis de estructuras uniformes y discontinuidades planares.
-el método del dominio espectral para el análisis de estructuras planares uniformes.
-el método del circuito planar con corrección de la dispersión.

Se ha elegido el método de la resonancia transversal generalizado como el mas idóneo para el análisis preciso y rápido de discontinuidades finline con un mínimo de recursos informáticos. Se ha desarrollado incluyendo el carácter singular de los campos en las aristas de la estructura, consiguiéndose minimizar el fenómeno de la convergencia relativa y obteniéndose resultados precisos con expansiones reducidas de las funciones modales que representan los campos, validándose los resultados obtenidos al analizar diversas estructuras con medidas experimentales.

This thesis investigates profiling and differentiating customers through the use of statistical data mining techniques. The business application of our work centres on examining individuals’ seldomly studied yet critical consumption behaviour over an extensive time period within the context of the wireless telecommunication industry; consumption behaviour (as oppose to purchasing behaviour) is behaviour that has been performed so frequently that it become habitual and involves minimal intentions or decision making. Key variables investigated are the activity initialised timestamp and cell tower location as well as the activity type and usage quantity (e.g., voice call with duration in seconds); and the research focuses are on customers’ spatial and temporal usage behaviour. The main methodological emphasis is on the development of clustering models based on Gaussian mixture models (GMMs) which are fitted with the use of the recently developed variational Bayesian (VB) method. VB is an efficient deterministic alternative to the popular but computationally demandingMarkov chainMonte Carlo (MCMC) methods. The standard VBGMMalgorithm is extended by allowing component splitting such that it is robust to initial parameter choices and can automatically and efficiently determine the number of components. The new algorithm we propose allows more effective modelling of individuals’ highly heterogeneous and spiky spatial usage behaviour, or more generally human mobility patterns; the term spiky describes data patterns with large areas of low probability mixed with small areas of high probability. Customers are then characterised and segmented based on the fitted GMM which corresponds to how each of them uses the products/services spatially in their daily lives; this is essentially their likely lifestyle and occupational traits. Other significant research contributions include fitting GMMs using VB to circular data i.e., the temporal usage behaviour, and developing clustering algorithms suitable for high dimensional data based on the use of VB-GMM.

Mnoho rozdílných strategií fúze bylo vyvinuto během posledních 15 let výzkumu simultánního EEG-fMRI. Aktuální dizertační práce shrnuje aktuální současný stav v oblasti výzkumu fúze simultánních EEG-fMRI dat a pokládá si za cíl vylepšit vizualizaci úkolem evokovaných mozkových sítí slepou analýzou přímo z nasnímaných dat. Dva rozdílné modely, které by to měly vylepšit, byly navrhnuty v předložené práci (tj. zobecněný spektrální heuristický model a zobecněný prostorovo-frekvenční heuristický model). Zobecněný frekvenční heuristický model využívá fluktuace relativního EEG výkonu v určitých frekvenčních pásmech zprůměrovaných přes elektrody zájmu a srovnává je se zpožděnými fluktuacemi BOLD signálů pomocí obecného lineárního modelu. Získané výsledky ukazují, že model zobrazuje několik na frekvenci závislých rozdílných úkolem evokovaných EEG-fMRI sítí. Model překonává přístup fluktuací absolutního EEG výkonu i klasický (povodní) heuristický přístup. Absolutní výkon vizualizoval s úkolem nesouvisející širokospektrální EEG-fMRI komponentu a klasický heuristický přístup nebyl senzitivní k vizualizaci s úkolem spřažené vizuální sítě, která byla pozorována pro relativní pásmo pro data vizuálního oddball experimentu. Pro EEG-fMRI data s úkolem sémantického rozhodování, frekvenční závislost nebyla ve finálních výsledcích tak evidentní, neboť všechna pásma zobrazily vizuální síť a nezobrazily aktivace v řečových centrech. Tyto výsledky byly pravděpodobně poškozeny artefaktem mrkání v EEG datech. Koeficienty vzájemné informace mezi rozdílnými EEG-fMRI statistickými parametrickými mapami ukázaly, že podobnosti napříč různými frekvenčními pásmy jsou obdobné napříč různými úkoly (tj. vizuální oddball a sémantické rozhodování). Navíc, koeficienty prokázaly, že průměrování napříč různými elektrodami zájmu nepřináší žádnou novou informaci do společné analýzy, tj. signál na jednom svodu je velmi rozmazaný signál z celého skalpu. Z těchto důvodů začalo být třeba lépe zakomponovat informace ze svodů do EEG-fMRI analýzy, a proto jsme navrhli více obecný prostorovo-frekvenční heuristický model a také jak ho odhadnout za pomoci prostorovo-frekvenční skupinové analýzy nezávislých komponent relativního výkonu EEG spektra. Získané výsledky ukazují, že prostorovo-frekvenční heuristický model vizualizuje statisticky nejvíce signifikantní s úkolem spřažené mozkové sítě (srovnáno s výsledky prostorovo-frekvenčních vzorů absolutního výkonu a s výsledky zobecněného frekvenčního heuristického modelu). Prostorovo-frekvenční heuristický model byl jediný, který zaznamenal s úkolem spřažené aktivace v řečových centrech na datech sémantického rozhodování. Mimo fúzi prostorovo-frekvenčních vzorů s fMRI daty, jsme testovali stabilitu odhadů prostorovo-frekvenčních vzorů napříč různými paradigmaty (tj. vizuální oddball, semantické rozhodování a resting-state) za pomoci k-means shlukovacího algoritmu. Dostali jsme 14 stabilních vzorů pro absolutní EEG výkon a 12 stabilních vzorů pro relativní EEG výkon. Ačkoliv 10 z těchto vzorů vypadají podobně napříč výkonovými typy, prostorovo-frekvenční vzory relativního výkonu (tj. vzory prostorovo-frekvenčního heuristického modelu) mají vyšší evidenci k úkolům.

The natural discontinuities in rock masses, which form unstable rock blocks, have a profound impact on the stability and safety of mining structures. The most commonly used means of rock block reinforcement in the field is fully grouted rock bolt because of its high tension capacity and its efficient anchoring. Rock bolting system forms a self-supporting structure in rock mass through reinforcing loosened rock blocks, improving shear strength of rock joints. According to field observations the failure of rock bolts occurs due to a combination of both pull-out and shear forces. Thus, understanding the failure mechanism of bolted rock joint under such a mixing loading condition is essential for rock support system design. The surface roughness characteristics, Constant Normal Load (CNL) and Constant Normal Stiffness (CNS) conditions, and the presence of infill material within rock joint can significantly influence its shear strength. Moreover, the mechanical and failure behaviour of rock as a heterogeneous material is controlled by various microstructural parameters, such as grain shape and size, type of minerals, and the existence of pre-existing flaws. Any damage due to the mine roof fall (e.g. rock block collapse in roadways and tunnels) or the failure of rock in open pit slopes can hinder mining activities, and results in penalties being imposed on mining companies. Therefore, an appropriate evaluation of rock block instability and response of rock bolting system is critical when designing both surface and underground mining structures. Recent developments in computational mechanics and distinct element numerical method (DEM) enable more efficient and faster design of mining structures. However, a promising DEM framework requires a robust and rigorous contact constitutive model, which is capable of mimicking the failure and mechanical response of material at microscopic scale. The key aspect of DEM contact model is its contact force-displacement law, which is responsible for capturing the essential macroscopic features of material failure and deformation. For rock joints reinforced with fully grouted rock bolts, these macroscopic features include brittle or softening behaviour of rock and grout, the cohesive or non-cohesive behaviour of infill material during shearing, and the failure of bolt-grout interface due to tension load. In the case of polycrystalline rock (e.g. granite), the inter- and intra-granular micro-cracking behaviour should also be taken into considerations. The focus of this study is on development of a DEM-based cohesive contact model for simulating the failure behaviour of rock, cohesive infill material (e.g. clay), grout, and bolt-grout interface. The proposed DEM-based cohesive model couples damage mechanics and plasticity theory in both modes I and II, and features an exponential decay damage function that considers the influence of both normal and shear stresses in reproducing a gradual, post-peak softening response in DEM contacts. Unlike conventional contact models such as Parallel Bond Model (PBM), flat-joint model (FJM), and smooth joint model (SJM), which feature no gradual degradation of contact strength after yield point, the cohesive softening behaviour incorporated in the new contact model inhibits the abrupt contact failure that enhances the macroscopic softening response of the DEM model. The proposed contact model is implemented in DEM code (PFC2D) to develop a cohesive DEM framework. A Stepwise Pull-Shear Test (SPST) scheme is developed to investigate the influence of pretension load, rib angle, and CNS boundary condition on the ultimate shear resistance of rock joints. The SPST approach allows simulation of bolted rock joints subjected to a combined pull-shear load, which is more realistic compared to previous shear testings that neglect the impact of simultaneous pull-out and shear loads. The proposed cohesive contact model is also incorporated into a Grain Based Model (GBM) to develop a cohesive GBM framework for simulating the micro-cracking behaviour of polycrystalline rocks. The numerical validations against a range of laboratory tests demonstrate that the proposed cohesive DEM and GBM frameworks are effective in reproducing the mechanical and failure behaviour of rock and grout materials as well as bolt-grout interface, the cohesive macroscopic response of clay-infilled rock joints, and micro-cracking behaviour of granitic rocks. The proposed modelling method, in conjunction with the SPST scheme, provided an efficient and inexpensive numerical framework that can be used by designers and geotechnical engineers for carrying out realistic experiments (i.e., combined pull–shear loads). Doing so will give them new insights into the mechanical performance of fully grouted rock bolts and failure behaviour of rock mass.
Thesis (Ph.D.) -- University of Adelaide, School of Civil, Environmental & Mining Engineering, 2019

碩士
華梵大學
資訊管理學系碩士班
98
Signal strength plays a key role in mobile communications. Therefore, accurate 2G base station positioning is an interesting area of study. In emergency calls, mobile communications affects the speed of response time. In aiming to avoid dropped-calls, accurate base station positioning is extremely important. Outliers inevitably occur when we collect data in the physical environment. These outliers produce bias when we estimate base station positions. Therefore, we identify outliers first for exclusion from our data. Then, based on TOA-based algorithms, we use K-means and random combination methods to estimate the location of base stations. The experiment results show that these methods can enhance the accuracy of base station positioning. In the course of the experiment, we originally removed the noise via a Support Vector Machine (SVM). But there is no way to identify data accuracy from the mixed data, as the error judgment of SVM affects the positioning of base stations. To rectify the lack of SVM, we clustered the data distribution to exclude noise data. In other words, the data required for accurate base station positioning falls on a particular region. Using outliers to estimate the base station locations results in estimates that are far from the correct region. This paper shows that it can be effective to exclude noise from the statistics to enhance the accuracy of estimated base station positions.

博士
國立中央大學
資訊工程學系
106
All drivers have their own distinct driving habits, and usually hold and operate the steering wheel differently in different driving scenarios. In this study, we proposed a novel Gaussian mixture model (GMM)-based method that can improve the traditional GMM in modeling driving behavior. This new method can be applied to build a better driver authentication or identification system based on the accelerometer and orientation sensor of a smartwatch. To demonstrate the feasibility of the proposed method, we created two experimental systems that analyzes driving behavior using the built-in sensors of a smartwatch. The experimental results for driver authentication—an equal error rate (EER) of 4.46% in the simulated environment and an EER of 11.35% in the real-driving environment—confirm the feasibility of this approach. For driver identification, the experimental results indicated that the proposed approach had identification rates of 87.16% in a simulated environment and 73.07% in a real-driving environment.

碩士
國立臺灣大學
流行病學研究所
91
The Haseman-Elston method (1972) is known for analyzing quantitative traits, which is proceeded by collecting the marker genotypes of sib pair and their parents and the traits of the sibs, and then constructs the relationship between the expectation of the squared trait difference (SQD) of sibs and the estimated proportion of alleles shared identical by descent (i.b.d.) at the marker locus. It was derived by Haseman and Elston (1972) that the expectation of the SQD is linear in the estimated proportion of alleles shared i.b.d. at the marker locus. The subsequent researches are not beyond the scope of regressing the SQD on the estimated proportion of genes i.b.d. at the marker locus to test if the slope is significantly negative; or performing a nonparametric analysis to examine if there is correlation between the SQD and the estimated proportion of marker genes i.b.d.; or even substituting other transformations for SQD. In this thesis, we focus on regression of the SQD on the estimated proportion of marker genes i.b.d. In Chapter 3, we classify the whole data into several strata by their marker informativeness and then calculate the conditional variance of the SQD, which is the inverse of the data reliability. Analyses based on the classical linear model are questionable when the conditional variances are not constant. Moreover, the distribution of the SQD is more right-skewed than the normal distribution. In Chapter 4, we apply the generalized linear model (GLM) for allowing nonnormal distribution. The algorithm solving the asymptotic maximum likelihood estimator (MLE) in GLM is iteratively reweighted least squares (IRLS), provided the inverse of the variance function as the weight then implement the weighted least squares until the estimate converges, so it also provides a solution for dealing with the problem mentioned in Chapter 3, that is, the heterogeneous error variance. In Chapter 5, we compare the GLM with the simple linear regression (SLR) by simulation. We find that the two methods perform similarly in large samples. The GLM approach improves the analyses of the Haseman-Elston method in small or moderate samples. We have used a diallelic marker for the simulation study, causing the data with complete marker i.b.d. information reduced. However, with the more informative molecular markers currently discovered, the number of marker alleles is expected to range from four to ten, which increases the portion of data with complete marker i.b.d. information and violates the assumption of constant error variance more severely. In that situation, using the GLM approach will be more meaningful.

碩士
國立中央大學
資訊工程學系
106
As the technology required for the Internet of Vehicles becomes more complete, the accompanying business opportunities and risks becomes higher. And the Driver Authentication and Identification will be an important issue in the future. Previous research proposed a driving behavior modeling method base on Gaussian mixture model. The method overcome the problem of traditional Gaussian mixture model in driving behavior modeling and proves the advantage on driver authentication. This paper will apply it to open-set driver identification. And use 10 mechanisms composed 3 extending modeling methods to discuss each effect on close-set registrant identification and registrant authentication. Finally, we experiment in the real-drive environment. And want to know that weather will get the same conclusion or not. The result shows that whatever modeling methods we used on close-set registrant identification, the EER can be reduced effectively as long as the method proposed by previous research is used on registrant authentication. The GMM’s EER is 23.182% and the method’s is 11.185% in simulated environment; The GMM’s EER is 33.657% and the method’s is 17.372% in real-drive environment.

Tese de doutoramento em Engenharia de Tecidos, Medicina Regenerativa e Células Estaminais
Regeneration, reconstruction and repair of peripheral nerve injuries (PNIs) are among the most complex and demanding challenges in the field of regenerative medicine. As a promising alternative to the “gold standard” autologous nerve grafts, tissue-engineered nerve guidance conduits (NGCs) have been extensively studied. However, in order to be able to produce an adequate NGC, the basic principles of neuro-biology must be known and followed. Also, great efforts have been made in terms of pre-clinical and clinical applications of engineered biomaterials for peripheral nerve regeneration. Furthermore, nanotechnology approaches have propelled nerve regenerative strategies. These themes are the focus of Section 1, in Chapters I, II and III. This PhD project has focused mainly on researching, developing and testing natural and biodegradable biomaterials using several fabrication methods to obtain the NGCs (Chapter IV). The first explored biomaterial was chitosan, in Chapter V, where the suitability of chitosan membranes with three different low degrees of acetylation was evaluated. Furthermore, to enhance the membranes’ cell adhesion and angiogenic properties, extracted human hair keratin was combined to the previous described membranes (Chapter VI). In Chapter VII, chitosan NGCs which are currently being used in in the clinical setting (Reaxon®), were used in combination with different gellan gum luminal fillers, mimicking the native Bands of Büngner. In Chapter VIII, a ground-breaking silk fibroin (SF) fabrication method for NGCs was established. SF NGCs were produced using a novel technology that comprises a horseradish peroxidase-mediated crosslinking system. Such innovative processing method resulted in the filling of an international patent. The excellent outcomes demonstrated by the enzymaticallycrosslinked method lead to an original research paper (Chapter IX). In line with the previous work, and because of its resourcefulness, a step forward was given, and a state-of-the-art study was performed, comprising the production of enzymatically-crosslinked SF NGCs capable of incorporating and delivering different neurotrophic factors. The developed NGCs restored the functional activity and enhanced neovascularization in the regenerated nerves, revealing comparable results to the autograft (Chapter X). In summary, the presented outcomes contributed to improve the actual state-of-art in what regards peripheral nerve regeneration, providing promising and versatile alternatives by using different types of natural-based biomaterials and approaches (Chapter XI).
A regeneração, reconstrução e reparação das lesões do nervo periférico encontram-se entre os desafios mais exigentes e complexos na área da Medicina Regenerativa. Como tratamento alternativo aos enxertos de nervo autólogo, Estruturas tubulares e protetoras do nervo produzidos, através de técnicas de Engenharia de Tecidos têm sido extensivamente estudados. Para além disso, grandes esforços têm sido feitos em termos de aplicações clínicas e pré-clínicas de biomateriais para a regeneração do nervo periférico. Abordagens nanotecnológicas têm também impulsionado o avanço nesta área. Estes temas são abordados na Secção 1, nos Capítulos I, II e III. Este projeto de Doutoramento focou-se principalmente na investigação, desenvolvimento e testes laboratoriais de várias formulações de biomateriais, combinados com diferentes técnicas de fabrico (Capítulo IV) de estruturas tubulares, usando apenas polímeros biodegradáveis conjugados com outras funcionalizações físicas ou biológicas. O quitosano foi o primeiro material investigado, no Capítulo V e VI, onde foram estudadas diferentes membranas de quitosano, assim como a melhoria das propriedades biológicas e angiogénicas destas, através da adição de queratina extraída de cabelo humano. No Capítulo VII, tubos de quitosano atualmente usados em Humanos (Reaxon®) foram combinados com diferentes formulações de Gellan Gum, produzidos para preencher o lúmen dos tubos, mimetizando assim as bandas de Bungner. No Capítulo VIII, um inovador método de produção de tubos condutores à base da Fibroína da Seda foi desenvolvido, resultando no registo de uma patente internacional. Esta metodologia envolve o uso de um sistema de reticulação enzimática com base na enzima peroxidase e o peróxido de hidrogénio (oxidante). Os excelentes resultados demonstrados através deste sistema de produção de tubos originaram um artigo original (Capítulo IX). Seguindo a mesma linha de trabalho, e devido á versatilidade da técnica, estudos inovadores foram desenvolvidos, onde fatores neurotróficos foram incorporados nos tubos de Fibroína da Seda, com o objetivo de restaurar a funcionalidade motora assim como a re-vascularização dos nervos regenerados (Capítulo X). Em suma, os resultados apresentados nesta tese indicam a versatilidade dos diferentes biomateriais e formulações testadas, assim como das diferentes técnicas, contribuindo deste modo para melhorar o atual estado da arte no que diz respeito à regeneração do nervo periférico (Capítulo XI).

Astrocytoma, the tumor of astrocytic origin, accounts for about 60 % of the primary brain tumors. As per World Health Organization grading system, astrocytoma is classified as circumscribed astrocytoma (Grade I; pilocytic astrocytoma) and diffusely infiltrating astrocytoma. Grade I tumor is biologically benign and can be cured by surgical resection of the tumor. The diffusely infiltrating astrocytoma is further subclassified into grade II/diffuse astrocytoma (DA), grade III/anaplastic astrocytoma (AA) and grade IV/glioblastoma (GBM). Aggressiveness of the disease increases as the tumor progresses from lower grade to higher grade. In particular, GBMs are the most malignant and aggressive human cancers. For a newly diagnosed GBM patient, the current treatment option is surgical resection of the tumor followed by radiation and temozolomide therapy. Despite the treatment is multimodal (surgery+radiation+temozolomide) the median survival of GBM patients remain very low at 14.6 months. Although numerous markers with potential utility in prognosis and treatment of GBMs have been reported, they are yet to be translated into clinical utility. Our knowledge of understanding the complete biology of GBMs needs further comprehensive studies towards the identification of markers with potential utility to prognose/treat the GBM patients efficiently. Therefore, with an immense need to develop new biomarkers/therapeutic strategies in order to improve the diagnosis, prognosis and existing treatment of the GBM, the current work is designed to study the following aspects on glioma

Surrogates of biodiversity are necessary tools for guiding the effective conservation of biodiversity. One of the best known approaches to assessing biodiversity is cross-taxonomic surrogacy, which is underpinned by the hypothesis that selected taxa (i.e. the surrogate) can provide useful and commensurate information on other components of biodiversity (i.e. the target). In this thesis, I examined the effectiveness of cross-taxonomic surrogacy by assessing cross-taxonomic patterns of congruence among ecologically important vertebrate and invertebrate taxa, and with respect to time and different landscape contexts. Using a long-term dataset, I first assessed patterns of cross-taxonomic congruence between three vertebrate groups over a 15-year period. My analyses revealed that patterns of cross-taxonomic congruency were inconsistent over time, varied among the taxa compared, and across different landscape contexts. Bird and mammal diversity were weakly concordant, but strengthened with time. However, there was little association between either birds or mammals, and reptiles. My findings suggested that cross-taxonomic surrogacy has limited effectiveness in heavily disturbed landscapes such as Nanangroe where ecological communities are expected to exhibit high temporal variation. Second, I examined the responses of two ecologically important insect groups (wild bees, beetles) to landscape context. Here, I found that species richness of bee assemblages showed no clear responses to different landscape contexts, unlike beetle assemblages. These patterns persisted even when both insect assemblage was partitioned into functionally-defined groups. Further analyses showed that both groups were responding to different landscape and vegetation components. My findings here demonstrated that wild bee diversity is weakly congruent with beetle diversity, and that surrogacy relationships between even charismatic insects should not be assumed without rigorous testing. Third, I examined how sets of woodland patches prioritised to best conserve each surrogate group (bird, herpetofauna, bee, beetle, tree) represented the other four groups using a complementarity-based approach. Thereafter, I compared these findings with correlation-based analysis to determine patterns of cross-taxonomic congruence. I found that patch sets selected to optimise representation of the surrogate varied in how it incidentally represented other taxa. Beetles achieved the highest incidental representation of other taxa while bees and trees performed the worst. Yet, beetles were the most costly taxa to conserve given the large number of patches needed to meet beetle targets, an ecological consequence of the high diversity and compositional turnover of beetle assemblages. My findings show that species diversity of any taxa should be a pertinent consideration in identifying cross-taxonomic surrogates to prioritise sites for biodiversity conservation. Fourth, I performed a meta-analytical review of the global surrogate literature to assess the effects of anthropogenic disturbance on cross-taxonomic surrogacy in terrestrial systems. Drawing from a dataset compiled from 146 studies, my analyses revealed that anthropogenic disturbance plays an important role in shaping patterns of cross-taxonomic congruence, especially at landscape and regional scales. Spatial scale was an important predictor of cross-taxonomic patterns, but only at very large scales. In conclusion, my findings caution against extrapolating cross-taxonomic surrogates across landscapes subjected to different levels of disturbance and spatial scales to assess biodiversity. Focusing on ubiquitous, human-modified landscapes, my work underscores a number of practical and theoretical issues concerning the use of cross-taxonomic surrogacy. By collectively or individually examining the roles of time, landscape context and habitat structure with respect to diverse groups of vertebrate and invertebrate taxa, my thesis makes explicit the need to consider important ecological processes that can better guide the use of biodiversity surrogates in conservation.

碩士
靜宜大學
英國語文學系
105
Grammar acquisition in English as foreign language (EFL) setting has been a challenging task for language learners, including Taiwan. Various pedagogies are utilized to explore EFL students’ grammar learning; for example, the Grammar Translation Method (GTM) among three instructional approaches, featuring drills as lecture-based instruction on grammar usages, has a longest teaching history and is still being extensively practiced. Afterwards, the Communicative Language Teaching (CLT) approach has shed light on the significance of interaction in order to boost communicative competence, even in grammar learning. More recently, the prevalence of computer assisted language learning (CALL) promoted the Flipped Classroom (FC) to become an innovative, effective teaching mode, where learners preview on-line instructional materials before class and then are engaged to interactive learning activities during face-to-face meetings. While previous studies have asserted the effects of respective pedagogy on language learning, research comparing the abovementioned three instructional approaches in language learning remains scarce, not to mention such a comparison in grammar acquisition among EFL learners. The present study, hence, aimed to investigate EFL junior high school students in central Taiwan on their grammar learning effects of diverse instructional approaches. Three groups of students received the same English grammar content including infinitive verbs, gerunds, indirect speech, causative verbs, verbs of the senses, linking verbs, and relative pronouns of distinct methods (GTM, CLT, and FC respectively) for a six-week teaching period. Multiple sources of data were collected, comprising grammar written and speaking pre-tests, three different ranges of achievement tests, grammar written and speaking post-tests, and semi-structured interviews. Based on the outcome of the written and speaking pre-tests, the three groups demonstrated similar grammar competence, and the results of the written post-tests indicated that the FC group performed significantly better than the other two groups; yet the significant difference was not found between the GTM group and CLT group. As for the speaking tests, both the GTM and FC groups improved significantly, with much more improvement in the FC class. These findings contribute to a better understanding of how FC produces the paramount learning impact than GTM and CLT. The analysis of the interview suggested that the students found the You-tube videos of FC approach particularly practical in facilitating them to learn grammar, thus promoting the following acquisition. Ultimately, pedagogical implications and suggestions for future research are provided in the conclusion.

The aftermath of the financial crisis that had its beginning in 2007 has put to the fore the increase of the public debt in OECD countries. Global economic recessions and public bailouts of banks have resulted in a significant concern about the sovereign default risk mainly on the Eurozone countries facing structural economic imbalances. Using Credit Default Swaps (CDS) as a measure of sovereign credit risk, the purpose of this study is to analyze the link between sovereign default risk and income inequality for a broad panel of 26 European countries over the years 2005 to 2010. Applying the System GMM techniques the findings support the hypothesis that income inequality is a significant predictor of the sovereign credit risk. The empirical results also show that income inequality has more impact on the dynamics of the CDS spreads in times of economic downturns.
A recente crise financeira que teve origem no início de 2007 expôs o aumento da dívida pública em alguns países da OCDE. Os desequilíbrios macroeconómicos e a vulnerabilidade do sistema financeiro global estão associados ao risco de falência soberano principalmente nos países da Zona Euro desprovidos de mecanismos autónomos de política monetária. O objetivo deste estudo é analisar a relação existente entre o risco de falência soberano medido através dos spreads dos CDS e a desigualdade de rendimento para um painel de 26 países Europeus ao longo do período de 2005 a 2010. Utilizando o estimador GMM-System como metodologia econométrica, os resultados do trabalho empírico suportam a hipótese de partida de que a desigualdade de rendimento constitui um importante determinante no risco de crédito soberano. Como principal evidência empírica a retirar deste estudo, sublinha-se que a desigualdade de rendimento tem maior impacto na dinâmica dos CDS em períodos de recessão económica.

Dissertação de mestrado integrado em Engenharia Biomédica (área de especialização em Eletrónica Médica)
Glioblastoma Multiforme (GBM) is a harmful brain tumor with a median overall survival (OS) of only fourteen months. Predicting the OS from pre-treatment Magnetic Resonance Imaging (MRI) is intrinsically tricky not only due to the morphologic characteristics of the tumor but also due to the impact of the treatment. In the last years, deep learning began to arouse the interest of the scientific community due to the excellent performances achieved. In the medical imaging field, the convolutional neural networks (CNN) are achieving better performances than the clinicians in several tasks. This dissertation aims to develop a deep learning model for regression to directly predict the OS using pre-treatment MRI scans from patients with glioblastoma. For this, three different approaches were used: the transference of knowledge from a 2D and a 3D CNN and training from scratch on a 3D CNN. The effects of inputting different MRI modalities with different preprocessing techniques, such as the region of interest (ROI) and z-score normalizations, were studied. After a first selection of the top-ranked models based on two distinct criteria, data augmentation was performed. Throughout this work, the approaches were analyzed individually and then compared with each other and with a state-of-the-art approach. Although it was not possible to conclude on the MRI modalities and z-score normalization that benefits the most the OS prediction, a preference for a broader context while training the model was evident. The importance maps from the 3D networks developed showed that the regions considered most important by the network overlap with the ones where the tumor is more frequent. The network with more potential for the OS prediction is the 3D CNN trained from scratch that used FLAIR and tumor segmentation as input, ROI as all image and z-score normalization to all the image. In the end, this model was not able to outperform the state of the art.
O Glioblastoma Multiforme é um tumor cerebral nocivo com uma sobrevida mediana de apenas catorze semanas. Prever o tempo de sobrevida a partir de imagens de ressonância magnética é intrinsecamente difícil, não apenas devido às características morfológicas, mas também devido ao impacto dos tratamentos. Durante os últimos anos, o deep learning começou a despertar o interesse da comunidade científica devido às excelentes performances obtidas. Na área da imagem médica, as redes neuronais convolucionais estão a obter melhores performances que os médicos em diversas tarefas. Esta dissertação tem como objetivo desenvolver um modelo de regressão de deep learning para prever o tempo de sobrevida através de ressonâncias magnéticas pré-tratamento de pacientes com glioblastoma. Para tal, três diferentes abordagens foram utilizadas: a transferência de conhecimento de uma rede neuronal convolucional 2D e de outra 3D e ainda treinar do zero uma rede neuronal convolucional 3D. Os efeitos de introduzir na rede diferentes modalidade de ressonância magnética com diferentes préprocessamentos, tal como a região de interesse considerada ou a normalização padrão efetuada, foram estudados. Após uma primeira seleção dos modelos com melhores métricas baseada em dois critérios distintos foi efetuado o aumento artificial dos dados. Durante este trabalho, as diferentes abordagens foram analisadas individualmente e posteriormente comparadas entre si e com o estado da arte. Embora não seja possível concluir sobre modalidade de ressonância magnética e o tipo de normalização padrão que beneficiam a previsão do tempo de sobrevida, e preferência por um contexto visual mais abrangente foi evidente. Os mapas de importância relativos às redes neuronais 3D mostram que as regiões consideradas mais importantes pela rede coincidem com as que o tumor é mais frequente. A rede que revelou um maior potencial para prever o tempo de sobrevida é a rede neuronal convolutional 3D treinada do zero que utiliza a sequência FLAIR conjugada com a segmentação do tumor como entrada da rede, a região de interesse é toda a imagem e a normalização padrão é efetuada a toda a sequência. Por fim, este modelo não foi capaz de ultrapassar o estado da arte.

Glioblastoma (GBM) is the most common high-grade brain tumor diagnosed in patients who are more than 50 years of age. The standard of care treatment is surgery, followed by radiotherapy and chemotherapy. The median life expectancy of patients is only between 12 to 15 months after receiving current treatment regimes. Hence, identification of new therapeutic compounds and gene targets are highly warranted. This thesis describes four interlinked studies to attain this goal. In study 1, we explored drug combination effects in a material of 41 patient-derived GBM cell (GC) cultures. Synergies between three compounds, pterostilbene, gefitinib, and sertraline, resulted in effective killing of GC and can be predicted by biomarkers. In study 2, we performed a large-scale screening of FDA approved compounds (n=1544) in a larger panel of GCs (n=106). By combining the large-scale drug response data with GCs genomics data, we built a novel computational model to predict the sensitivity of each compound for a given GC. A notable finding was that GCs respond very differently to proteasome inhibitors in both in-vitro and in-vivo. In study 3, we explored new gene targets by RNAi (n=1112) in a panel of GC cells. We found that loss of transcription factor ZBTB16/PLZF inhibits GC cell viability, proliferation, migration, and invasion. These effects were due to downregulation of c-MYC and Cyclin B1 after the treatment. In study 4, we tested the genomic stability of three GCs upon multiple passaging. Using molecular and mathematical analyses, we showed that the GCs undergo both systematic adaptations and sequential clonal takeovers. Such changes tend to affect a broad spectrum of pathways. Therefore, a systematic analysis of cell culture stability will be essential to make use of primary cells for translational oncology. Taken together, these studies deepen our knowledge of the weak points of GBM and provide several targets and biomarkers for further investigation. The work in this thesis can potentially facilitate the development of targeted therapies and result in more accurate tools for patient diagnostics and stratification.

Mycobacterium tuberculosis, the etiological agent of tuberculosis, has adapted with the host environment and evolved to survive in harsh conditions in the host. The pathogen has successfully evolved strategies not only to evade the host immune system but also to thrive within the host cells. Upon infection, the pathogen is either cleared due to the host immune response, or it survives and causes active tuberculosis (TB) infection. In a number of cases however, the pathogen is neither killed nor does it actively proliferate, but it remains dormant in the host until the environment becomes favorable. This dormant state of pathogen is responsible for latent TB infection (LTBI). WHO reports indicated that as much as a third of the whole world’s population is exposed to the pathogen, of which a significant proportion could be latently infected (WHO report, 2015). These individuals do not show symptoms of active TB infection and hence are difficult to detect. The latent TB infected (LTBI) individuals serve as a reservoir for the pathogen, which can lead to epidemics when the conditions change. Hence, it is necessary to understand the host -pathogen interactions during LTBI, as this might provide clues to developing new strategies to detect and curb a latent infection. Host-pathogen interactions are multifaceted, in which both species attempt to recognize and respond to each other, all of these through specific molecules making distinct interactions with the other species. The outcome of the infection is thus decided by a complex set of host-pathogen interactions. The complexity arises since a large number of molecular components are involved, also multiplicity of interactions among these components and due to several feedback, feed forwards or other regulatory or influential loops within the system. The complexity of biological systems makes modeling and simulation an essential and critical part of systems– level studies. Systems biology studies provide an integrated framework to analyze and understand the function of biological systems. This work addresses some of these issues with an unbiased systems-level analysis so as to identify and understand the important global changes both in the host and in the pathogen during LTBI. The broad objectives of the work was to identify the key processes that vary in the host during latent infection, the set of metabolic reactions in the host which can be modulated to control the reactivation of infection, global adaptation in Mycobacterium tuberculosis (Mtb) and then to utilize this knowledge to identify strategies for tackling latent infection. A review of literature of the current understanding of latency from the pathogen and the host perspective is described in chapter 1. From this, it is clear that most available studies have focused on the role of individual molecules and individual biological processes such as granuloma formation, toll-like receptor signaling, T cell responses as well as cytokine signaling, in either initiating or maintaining a latent infection, but there is no report till date about whether and how these processes are connected with each other. While transcriptome based studies have identified lists of differentially expressed genes in LTBI as compared to healthy controls, no further understanding is currently available for many of them, regarding the processes they may be involved in and what interactions they make, which may be important for understanding LTBI. The first part of the work is a systematic meta-analysis of genome-scale protein interaction networks rendered condition-specific with transcriptome data of patients with LTBI, which has provided a global unbiased picture of the transcriptional and metabolic variations in the host and in the pathogen during the latent infection. To start with, publicly available gene expression data related to LTBI, active TB and healthy controls were considered. In all, 183 datasets summing up to 105 LTBI, 41 active TB and 37 healthy control samples were analyzed. (Chapter 2). Standard analysis of the transcriptome profiles of these datasets indicated that there was zero overlap among them and that not a single gene was seen in common among all datasets for the same condition. An extensive human protein-protein interaction network was constructed using information available from multiple resources that comprehensively contained structural or physical interactions and genetic interactions or functional influences. Nodes in this network represented individual proteins and edges represented interactions between pairs of nodes. The identity of each node and the nature of interaction of each edge along with the type of evidence that was used as the basis for drawing the edge, was collated for the network. The gene expression data was integrated into the human protein-protein interaction (PPI) network for each condition, which essentially had weighted nodes and directed edges, specific to that condition, from which specific comparative networks were derived. The highest ranked perturbations in LTBI were identified through a network mining protocol previously established in the laboratory. This involved computing all versus all shortest paths on the comparative network, scoring the paths based on connectedness and various centrality measures of the nodes and the edges and finally ranking the paths based on the cumulative path scores. Intriguingly, the top-ranked set of perturbations were found to form a connected sub-network by themselves, referred to as a top perturbed sub-network (top-net), indicating that they were functionally linked or perhaps even orchestrated in some sense. Th17 signaling appears to be dominant. About 40 genes were identified in the unique set of LTBI condition as compared to the active TB condition, and these genes showed enrichment for processes such as apoptosis, cell cycle as well as natural killer cell mediated toxicity. Construction and analysis of a miRNA network indicated that 32 of these have strong associations with miRNA explaining the role of the latter in controlling LTBI. 3 other genes from the top-net are already established drug targets for different diseases with known drugs associated with them, which are BCL2, HSP90AA1 and NR3C1. These 3 proteins can be explored further as drug targets in LTBI whose manipulation using existing drugs may result in inhibiting the underlying biological process and thereby result in disturbing the state of latency. As a second objective, global variations in the host transcriptome were identified during ascorbic acid induced dormancy (Chapter 3). Ascorbic acid or Vitamin C is a nutrient supplement required in the diet. This organic compound has a known antioxidant property, as it is known to scavenge the free radicals. In a recent study, Taneja et al, demonstrated that Vitamin C could induce dormancy in Mtb. On similar lines, experiments were done in THP-1 cells infected with Mtb to determine the host responses during ascorbic acid (AA) induced dormancy. The raw gene expression data was provided by our collaborator Prof. Jaya Tyagi that included 0 hour, 4 days and 6 days time points with infection and vitamin C versus infection alone or vitamin C alone as controls. The transcriptome data was normalized and integrated into the human PPI network as described for the meta-analyses. It was experimentally determined that ascorbic acid induces dormancy in 4 days post infection. The top-ranked paths of perturbation were analyzed and compared for three different conditions: (i) uninfected condition, (ii) AA treated and infected condition, and (iii) AA, isoniazid and infected condition. The dormant pathogen is known to be drug-tolerant and thus as a marker for the state of dormancy, the lack of effect of isoniazid is also monitored in the infected host cells. The analysis revealed that there were some broad similarities as compared to LTBI from patient samples but AA induced dormancy in cell lines stood out a separate group indicating that there were significant differences such as involving Interferon Induced Transmembrane Proteins (IFITMs), vacuolar ATPase as well as GDF15, which belongs to TGF-beta signaling pathway. The highest ranked perturbed paths contained genes involved in innate immune responses of which ISG15, IFITMs, HLAs and ATPases emerge as the most altered in the dormant condition. CCR7 emerges as a key discriminator, which is subdued in the latent samples but highly induced in infection conditions. Pathway-based analysis of different conditions showed that oxidative stress, glutathione metabolism, proteasome degradation as well as type II interferon signaling are significantly up-regulated in AA induced dormancy. The dormant bacteria reside in the host cells and are known to modulate the host metabolism for their own benefit. So, the third objective was to understand the metabolic variations in the host during LTBI (Chapter 4). A genome-scale metabolic (GSM) model of alveolar macrophage was used in this study. The metabolic model contains information of the reactions, metabolites and the genes encoding enzymes that catalyze a particular reaction. Flux balance analysis (FBA), a constraint-based metabolic modeling method, is used for analyzing the alterations in the metabolism under different infection conditions. In order to mimic the physiological condition, gene expression data was used for constraining the bounds of the reactions in the model. Two different expression studies were used for analysis: GSE25534 (from Chapter 2) and ascorbic acid induced dormancy (Chapter 3). The analysis was carried out for latent TB versus healthy control and latent TB versus active TB to identify the most altered metabolic processes in LTBI. Differences in fluxes between the two conditions were calculated. A new classification scheme was devised to categorize the reactions on the basis of flux differences. In this chapter, higher fluxes in LTBI condition were identified for reactions involved in transport of small metabolites as well as amino acids. Solute carrier proteins responsible for the transport of the metabolites were identified and their biological significance is discussed. Reduced glutathione (GSH), arachidonic acid, prostaglandins, pantothenate were identified as important metabolites in LTBI condition and their physiological role has been described. Sub-system analysis for different conditions shows differential regulation for arachidonic acid metabolism, fatty acid metabolism, folate metabolism, pyruvate metabolism, glutathione metabolism, ROS detoxification, triacylglycerol synthesis and transport as well as tryptophan metabolism. From the study, transporter proteins and reactions altered during LTBI were identified, which again provide clues for understanding the molecular basis of establishing a latent infection. Mycoabcterium tuberculosis is known to undergo dormancy during stress conditions. In this chapter, the main objective was to identify the global variations in the dormant Mtb (Chapter 5). To carry out the analysis, the Mtb PPI network was constructed using information from available resources. Gene expression data of two different dormancy models, Wayne growth model and multiple-stress model, were used for the study. To identify the key players involved in reversal of dormancy, the transcriptome data of reaeration condition was also used. In this study, the Max-flow algorithm was implemented to identify the feasible paths or flows in different condition. The flows with higher scores indicate that more information is traversed by the path, and hence is important for the study. From the analysis of Wayne growth model (hypoxia model), important transcriptional regulators such as SigB, SigE, SigH, regulators in the two-component system such as MprA, MtrA, PhoP, RegX3 and TrcR were identified in stress condition. Multiple-stress model studied the growth of bacteria in low oxygen concentration, high carbon dioxide levels, low pH and nutrient starvation. The gene expression data was integrated in the Mtb PPI network and implementation of Max-flow algorithm showed that MprA, part of the MprA-MprB two-component system, is involved in the regulation of persistent condition. WhiB1 also features in the paths of dormant condition and its role in persistence can be explored. In reaeration model, WhiB1 and WhiB4 are present in the top flows of this condition indicating that the redox state is perturbed in the pathogen and the interactions of these proteins are important to understand the reversal of dormant condition. From the study, Rv2034, Rv2035, HigA, Rv1989, Rv1990 and Rv0837 proteins belonging to toxin-antitoxin systems were also identified in the dormant bacteria, indicating their role in adaptation during stress condition. The role of Rv2034 has been studied in persistence, but the function of other proteins can be analyzed to provide new testable hypotheses about the role of these proteins in dormancy. Thus, the flows or paths perturbed during dormancy were identified in this study. To get a better understanding of the metabolic network active in mycobacteria under different conditions, experiments were performed in Mycobacterium smegmatis MC2 155. The non-pathogenic strain of genus Mycobacteria, Mycobacterium smegmatis, is used as a surrogate to carry out molecular biology studies of Mtb. Mycobacterium smegmatis MC2 155 (Msm) is the commonly used laboratory strain for experimental purpose. In order to obtain a clear understanding of how comparable are the metabolic networks between the virulent M. tuberculosis H37Rv and the model system Msm, the latter model is first studied systematically. In Chapter 6, first the functional annotation of the Msm genome was carried out and the genes were categorized into different Tuberculist classes based on homology with the Mtb genome. A high-throughput growth characterization was carried out to characterize the strain systematically in terms of different carbon, nitrogen or other sources that promoted growth and thus served as nutrients and those that did not, together yielding a genome-phenome correlation in Msm. Gene expression was measured and used for explaining the observed phenotypic behavior of the organism. Together with the genome sequence, the transcriptome and phenome analysis, a set of about 257 different metabolic pathways were identified to be feasible in wild-type Msm. About 284 different carbon, nitrogen source and nutrient supplements were tested in this experiment and 167 of them supported growth of Msm. This indicates that the compounds enter the cells and are metabolized efficiently, thus yielding similar phenotypes. The expressed genes and metabolites supporting growth were mapped to the metabolic network of Msm, thus helping in the identification of feasible metabolic routes in Msm. A comparative study between Msm and Mtb revealed that these organisms share similarity in the nutrient sources that are utilized for growth. The study provides experimental proof to identify the feasible metabolic routes in Msm, and this can be used for understanding the metabolic capability in the two organisms under different conditions providing a basis to understand adaptations during dormancy. In the last part of the work presented in this thesis, the metabolic shift in the pathogen was studied using a genome-scale metabolic model of Mtb (Chapter 7). The model contains information of the reactions, metabolites and genes involved in the reactions. Flux balance analysis (FBA) was carried out by integrating normalized gene expression data (Wayne model and multiple-stress model transcriptome considered in Chapter 5) to identify the set of reactions, which have a higher flux in the dormant condition as compared to the control replicating condition. Glutamate metabolism along with propionyl CoA metabolism emerge as major up-regulated processes in dormant Mtb. Next, with an objective of identifying essential genes in dormant Mtb, a systematic in silico single gene knock-out analysis was carried out where each gene and it's associated reaction was knocked out of the model, one at a time and the ability of the model to reach its objective function assessed. About 168 common genes in Wayne model and multiple-stress model were identified as important in Mtb after the knockout analysis. Essentiality is in essence a systems property and requires to be probed through multiple angles. Towards this, essential genes were identified in Mtb using a multi-level multi-scale systems biology approach. About 283 genes were identified as essential on the basis of combined analysis of transcriptome data, FBA, network analysis and phyletic retention studies in Mtb. 168 genes identified as important in dormant Mtb were compared with 283 essential genes and about 91 genes were found to be essential. Finally, among the set of essential genes, those that satisfy other criteria for a drug target were analyzed using the list of high-confidence drug targets of Mtb available in the laboratory along with their associated drug or drug-like molecules. 38 out of the 168 important genes in Mtb were found to have one or more drugs associated with them from the DrugBank database. Colchicin-Rv1655, Raloxifene-Rv1653, Bexarotene-Rv3804, Rosiglitazone-Rv3804 are top-scoring drug-target pairs that can be explored for killing dormant bacilli. The study has thus been useful in identifying important proteins, reactions and drug targets in dormant Mtb. In summary, the thesis presents a comprehensive systems-level understanding of various aspects of host responses and pathogen adaptation during latent TB infection. Key host and pathogen factors involved in LTBI are identified that serve as useful pointers for deriving strategies for tackling a latent infection.

Glycosphingolipid (GSL) metabolism is a complex process involving proteins and enzymes at distinct locations within the cell. Mammalian GSLs are typically based on glucose or galactose, forming glucosylceramide (GlcCer) and galactosylceramide (GalCer). Most GSLs are derived from GlcCer, which is synthesized on the cytosolic leaflet of the Golgi, while all subsequent GSLs are synthesized on the lumenal side. We have utilized both pharamacological and genetic manipulation approaches to selectively regulate GSL metabolism and better understand its mechanistic details. We have developed analogues of GlcCer and GalCer by substituting the fatty acid moiety with an adamanatane frame. The resulting adamantylGSLs are more water-soluble than their natural counterparts. These analogues selectively interfere with GSL metabolism at particular points within the metabolic pathway. At 40 µM, adaGlcCer prevents synthesis of all GSLs downstream of GlcCer, while also elevating GlcCer levels, by inhibiting lactosylceramide (LacCer) synthase and glucocerebrosidase, respectively. AdaGalCer specifically reduces synthesis of globotriaosylceramide (Gb3) and downstream globo-series GSLs. AdaGalCer also increases Gaucher disease N370S glucocerebrosidase expression, lysosomal localization and activity. AdaGSLs, therefore, have potential as novel therapeutic agents in diseases characterized by GSL anomalies and as tools to study the effects of GSL modulation. Two predominant theories have been developed to explain how GlcCer accesses the Golgi lumen: one involving direct translocation from the cytosolic-to-lumenal leaflet of the Golgi by the ABC transporter P-glycoprotein (P-gp, ABCB1, MDR1), and the other involving retrograde transport of GlcCer by FAPP2 to the ER, followed by entry into the vesicular transport system for Golgi lumenal access. To examine the in vivo involvement of P-gp in GSL metabolism, we generated a knockout model by crossbreeding the Fabry disease mouse with the P-gp knockout mouse. HPLC analyses of tissue Gb3 levels revealed a tissue-specific reduction in MDR1/Fabry mice. TLC analyses, however, did not show such reduction. In addition, we performed a gene knockdown study using siRNA against P-gp and FAPP2. Results show these siRNA to have distinct effects on GSL levels that are cell-type specific. These results give rise to the prospect of unique therapeutic approaches by targeting P-gp or FAPP2 for synthesis inhibition of particular GSL pathways.

Чепішко О. І. Сучасні підходи до реабілітації спортсменів ігрових видів спорту після травм гомілковостопного суглобу : кваліфікаційна робота магістра спеціальності 227 «Фізична терапія, ерготерапія» / наук. керівник О. В. Бессарабова. Запоріжжя : ЗНУ, 2020. 64 с.
UA : Кваліфікаційна робота – 64 сторінок, 9 таблиць, 2 рисунка, 70 літературних джерел. Об’єкт дослідження: процес відновлення спортсменів ігрових видів спорту після травм гомілковостопного суглобу. Мeтaдoслiджeння – вдосконалити програму фізичної реабілітації для спортсменів ігрових видів спорту з переломами гомілковостопного суглоба на різних етапах реконвалесценції. Мeтoдидoслiджeння: тeoрeтичнi: aнaлiз, клaсифiкaцiя, пoрiвняння, систeмaтизaцiя, узaгaльнeннятeoрeтичнихтaeмпiричнихдaних з прoблeми; Сeрeдeмпiричнихмeтoдiв – аналіз медичних карт, антропометрія, подометрія, констатуючий і формуючий етапи дослідження, спостереження; мeтoдимaтeмaтичнoїстaтистики (описова статистика, t-критерій Стьюдента). В роботі oбґрунтовaнo комплекс лікувальної гімнастики для спортсменів ігрових видів спорту з переломами ділянки гомілковостопного суглоба. Дoслiджeнooсoбливoстi клінічної картини на різних етапах реконвалесценції, а також періоди фізичної реабілітації на стаціонарному етапі. В роботі рoзширeнo, утoчнeнoтa деталізовано застосування засобів і методів оздоровчої фізичної культури на різних етапах реконвалесценції. В ході дослідження підтверджено, що розроблена програма реабілітації із застосуванням засобів фізичної терапії для спортсменів з травмами гомілковостопного суглоба забезпечує хороший терапевтичний ефект, дозволяє відмовитися від засобів додаткової опори в більш ранні терміни, сприяє відновленню функції гомілковостопного суглоба і працездатності пацієнта в залежності від локалізації перелому.
EN : Qualification work – 64 pages, 9 tables, 2 figures, 70 references. Object of research: the process of recovery of athletes in game sports after ankle injuries. The purpose of the study is to improve the program of physical rehabilitation for athletes of game sports with ankle fractures at different stages of convalescence. Research methods: theoretical: analysis, classification, comparison, systematization, generalization of theoretical and empirical data from the problem; among the empirical methods – analysis of medical records, anthropometry, podometry, ascertaining and forming the stages of research, observation; methods of mathematical statistics (descriptive statistics, Student’s t-test). The complex of medical gymnastics for sportsmen of sports with fractures of the ankle joint is substantiated in the work. The peculiarities of the clinical picture at different stages of convalescence, as well as the periods of physical rehabilitation at the inpatient stage have been studied. The paper expands, clarifies and details the application of means and methods of health-improving physical culture at different stages of convalescence. The study confirmed that the developed rehabilitation program using physical therapy for athletes with ankle injuries provides a good therapeutic effect, allows you to abandon the means of additional support at an earlier date, helps to restore ankle function and the patient’s ability to work.

This study analysed the effects of international trade on human development in two developing regions, the Association of Southeast Asian Nations (ASEAN) and the Southern African Development Community (SADC). The choice of comparing SADC and ASEAN is motivated by the many similarities between both regions half a century ago, and the stark divergence of their respective development pathways which has led to different development outcomes half a century later. Annual data from 2000 to 2018 and dynamic panel data econometric techniques were used in this study, controlling for individual country characteristics, endogeneity, serial correlation, heterocedasticity and interdependencies between the countries in each region. Two estimations were done in this study; sample wide estimations and country specific estimations. In the sample wide estimations the Generalised Method of Moments of Arellano and Bover (1995) with forward orthogonal deviations, and Feasible Generalised Least Squares of Parks (1967) and Kmenta (1986) were used, whilst Swamy’s Random Coefficients were used in the country specific estimations. Trade is measured using the current account balance as a percentage of GDP, whilst human development is captured by the United Nations’ Human Development Index (HDI). In the sample wide estimations, the study found that trade openness enhances human development for both SADC and ASEAN as measured by the Human Development Index (HDI). Gross fixed capital formation, economic growth and technological progress all had positive effects on human development in both regions. Unemployment had a counter intuitive positive effect on human development. This raises issues on the nature and quality of employment, including concerns on cheap production labour and vulnerable employment. The ASEAN region had a higher mean level of economic growth, a trade surplus and higher level of technological progress than SADC. This is consistent with the manufacturing focus of ASEAN, compared to the primary commodity exporting nature of SADC which had a trade deficit. However, in each region there were country specific differences in terms of what drives human development. The country specific disparities in drivers of human development have implications for the regional trade and development nexus. In particular, these disparities must be considered in the conceptualization and implementation of the SADC Industrialisation and Strategy Roadmap, and the most recent African Continental Free Trade Area. The policy implication is that such regional trade agreements should accommodate countries’ specific heterogeneity as the policy pathways will differ between countries.
Business Management
D. Phil. (Management Studies)

A doença de Gaucher é uma doença rara, autossómica recessiva, que resulta de uma acumulação anormal de uma substância lipídica, conhecida como glucocerebrosídeo ou glucosilceramida. Normalmente, o glucocerebrosídeo é metabolizado pela enzima glucocerebrosidase (ou β-glucosidase ácida). São as mutações no gene da glucocerebrosidase 1, a causa mais prevalente de todas as formas da doença de Gaucher. Essa alteração genética significa que a enzima não funciona corretamente ou está totalmente ausente. Isso, por sua vez, leva à acumulação de glucocerebrosídeo nas células. É por causa desta aglomeração que a doença de Gaucher é chamada de distúrbio de armazenamento e, como o acúmulo ocorre no lisossoma, é uma desordem de armazenamento lisossomal (também conhecida como doença de armazenamento lisossomal, ou esfingolipidose), em que esse excesso de lípidos acontece nos macrófagos. A acumulação de lípidos nas células na doença de Gaucher é diferente para cada pessoa, o que acarreta uma desigual sintomatologia que varia consideravelmente de paciente para paciente. Algumas pessoas apresentam sintomas graves na infância, enquanto que outras não apresentam sintomas ou apresentam apenas sintomas leves e são diagnosticadas numa fase mais tardia da vida. Pacientes com doença de Gaucher podem manifestar danos neurológicos. Três tipos principais da doença de Gaucher têm vindo a ser descritos ao longo dos tempos, com base na ausência (tipo 1), presença (tipo 2 e tipo 3) ou da gravidade das características neurológicas. Um fenómeno multissistémico complexo surge envolvendo o fígado, baço, medula óssea e, ocasionalmente, os pulmões, havendo ainda casos de doença neurodegenerativa progressiva. O tratamento de reposição enzimática tornou-se o padrão de terapia em 1991, transformando a história natural da doença de Gaucher. Atualmente, tratamentos para doença de Gaucher, clinicamente usados para o tipo sistémico da doença, incluem a terapia enzimática de substituição, a terapia de redução de substrato e o transplante de células-tronco hematopoéticas, embora a última seja a única opção de sucesso para os tipos neuronopáticos da Doença de Gaucher. Este trabalho tem como objetivo fazer uma revisão narrativa da doença de Gaucher, contemplando, em particular, o tratamento usado.
Gaucher disease is a rare, autosomal recessive disease that results from an abnormal accumulation of a lipid substance known as glucocerebroside or glucosylceramide. Normally, glucoderebroside is metabolized by the enzyme glucocerebrosidase (or acid β-glucosidase). The mutations in the glucocerebrosidase 1 gene, are the most prevalent cause of all forms of Gaucher disease. This genetic alteration means that the enzyme does not work properly or is completely absent. This, in turn, leads to the accumulation of glucocerebroside in cells. It is because of this clumping that Gaucher disease is called a storage disorder and, as the accumulation occurs in the lysosome, it is a lysosomal storage disorder (also know as lysosomal storage disease, or sphingolipidosis) in which this excess of lipids happens in macrophages. The accumulation of lipids in cells in Gaucher disease is different for each person, resulting in unequal symptomatology that varies considerably from patient to patient. Some people have severe symptoms in childhood, while others have no symptoms or only mild symptoms and are diagnosed later in life. Gaucher disease patients may manifest neurological damage, three main types of Gaucher disease have been described over time, based on the absence (type 1), presence (type 2 and type 3) or the severity of neurological characteristics. A complex multisystem phenomenon arises involving the liver, spleen, bone marrow and, occasionally, the lungs, with cases of progressive neurodegenerative disease still occurring. Enzyme replacement treatment became the standard of therapy in 1991, transforming the natural history of Gaucher disease. Currently, treatments for Gaucher disease, clinically used for the systemic type of disease, include enzyme replacement therapy, substrate reduction therapy, and hematopoietic stem cell transplantation, although the latter is the only successful option for patients with the neuronopathic types of Gaucher disease. This work aims to make a narrative review of Gaucher disease, considering, in particular, the treatment used.