Statistical Hurdle Models for Single Cell Gene Expression: Differential Expression and Graphical Modeling

Thesis (Ph.D.)--University of Washington, 2016-06

Modelagem e otimização por metodologia de superfícies de resposta: um estudo em arames de aço SAE 9254 para molas automobilísticas

Pós-graduação em Engenharia Mecânica - FEG

Probabilistic Graphical Modelling for Software Product Lines: A Frameweork for Modeling and Reasoning under Uncertainty

This work provides a holistic investigation into the realm of feature modeling within software product lines. The work presented identifies limitations and challenges within the current feature modeling approaches. Those limitations include, but not limited to, the dearth of satisfactory cognitive presentation, inconveniency in scalable systems, inflexibility in adapting changes, nonexistence of predictability of models behavior, as well as the lack of probabilistic quantification of model’s implications and decision support for reasoning under uncertainty. The work in this thesis addresses these challenges by proposing a series of solutions. The first solution is the construction of a Bayesian Belief Feature Model, which is a novel modeling approach capable of quantifying the uncertainty measures in model parameters by a means of incorporating probabilistic modeling with a conventional modeling approach. The Bayesian Belief feature model presents a new enhanced feature modeling approach in terms of truth quantification and visual expressiveness. The second solution takes into consideration the unclear support for the reasoning under the uncertainty process, and the challenging constraint satisfaction problem in software product lines. This has been done through the development of a mathematical reasoner, which was designed to satisfy the model constraints by considering probability weight for all involved parameters and quantify the actual implications of the problem constraints. The developed Uncertain Constraint Satisfaction Problem approach has been tested and validated through a set of designated experiments. Profoundly stating, the main contributions of this thesis include the following: • Develop a framework for probabilistic graphical modeling to build the purported Bayesian belief feature model. • Extend the model to enhance visual expressiveness throughout the integration of colour degree variation; in which the colour varies with respect to the predefined probabilistic weights. • Enhance the constraints satisfaction problem by the uncertainty measuring of the parameters truth assumption. • Validate the developed approach against different experimental settings to determine its functionality and performance.

Développement et évaluation d’un environnement informatisé d’apprentissage pour faciliter l’intégration des sciences et de la technologie

Par cette recherche, nous voulons évaluer de manière exhaustive les bénéfices qu’apporte l’ExAO (Expérimentation Assistée par Ordinateur) dans les laboratoires scolaires de sciences et technologie au Liban. Nous aimerions aussi qu’elle contribue d’une manière tangible aux recherches du laboratoire de Robotique Pédagogique de l’Université de Montréal, notamment dans le développement du µlaboratoire ExAO. Nous avons voulu tester les capacités de l’ExAO, son utilisation en situation de classe comme : 1. Substitut d’un laboratoire traditionnel dans l’utilisation de la méthode expérimentale; 2. Outil d’investigation scientifique; 3. Outil d’intégration des sciences expérimentales et des mathématiques; 4. Outil d’intégration des sciences expérimentales, des mathématiques et de la technologie dans un apprentissage technoscientifique; Pour ce faire, nous avons mobilisé 13 groupe-classes de niveaux complémentaire et secondaire, provenant de 10 écoles libanaises. Nous avons désigné leurs enseignants pour expérimenter eux-mêmes avec leurs étudiants afin d’évaluer, de manière plus réaliste les avantages d’implanter ce micro laboratoire informatisé à l’école. Les différentes mise à l’essai, évaluées à l’aide des résultats des activités d’apprentissage réalisées par les étudiants, de leurs réponses à un questionnaire et des commentaires des enseignants, nous montrent que : 1. La substitution d’un laboratoire traditionnel par un µlaboratoire ExAO ne semble pas poser de problème; dix minutes ont suffi aux étudiants pour se familiariser avec cet environnement, mentionnant que la rapidité avec laquelle les données étaient représentées sous forme graphique était plus productive. 2. Pour l’investigation d’un phénomène physique, la convivialité du didacticiel associée à la capacité d’amplifier le phénomène avant de le représenter graphiquement a permis aux étudiants de concevoir et de mettre en œuvre rapidement et de manière autonome, une expérimentation permettant de vérifier leur prédiction. 3. L’intégration des mathématiques dans une démarche expérimentale permet d’appréhender plus rapidement le phénomène. De plus, elle donne un sens aux représentations graphiques et algébriques, à l’avis des enseignants, permettant d’utiliser celle-ci comme outil cognitif pour interpréter le phénomène. 4. La démarche réalisée par les étudiants pour concevoir et construire un objet technologique, nous a montré que cette activité a été réalisée facilement par l’utilisation des capteurs universels et des amplificateurs à décalage de l’outil de modélisation graphique ainsi que la capacité du didacticiel à transformer toute variable mesurée par une autre variable (par exemple la variation de résistance en variation de température, …). Cette activité didactique nous montre que les étudiants n’ont eu aucune difficulté à intégrer dans une même activité d’apprentissage les mathématiques, les sciences expérimentales et la technologie, afin de concevoir et réaliser un objet technologique fonctionnel. µlaboratoire ExAO, en offrant de nouvelles possibilités didactiques, comme la capacité de concevoir, réaliser et valider un objet technologique, de disposer pour ce faire, des capacités nouvelles pour amplifier les mesures, modéliser les phénomènes physiques, créer de nouveaux capteurs, est un ajout important aux expériences actuellement réalisées en ExAO.

Avaliação da influência da angulação dos implantes com diferentes tipos de conexão protética na distribuição de tensões: estudo pelo método dos elementos finitos tridimensionais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Influência da carga oclusal parafuncional em próteses implantossuportadas, variando-se a proporção coroa/implante e o sistema de conexão protética: estudo pelo método dos elementos finitos tridimensionais

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Análise das tensões em próteses unitárias implantossuportadas com diferentes tipos de conexões e qualidade óssea: estudo pelo método dos elementos finitos tridimensionais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Avaliação da distribuição das tensões em prótese do tipo “All on Four”: estudo pelo método dos elementos finitos tridimensionais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Likelihood-Based Inference for Partially Observed Multi-Type Markov Branching Processes

Thesis (Ph.D.)--University of Washington, 2016-08

Structure learning of context-specific graphical models

Abstract The ultimate problem considered in this thesis is modeling a high-dimensional joint distribution over a set of discrete variables. For this purpose, we consider classes of context-specific graphical models and the main emphasis is on learning the structure of such models from data. Traditional graphical models compactly represent a joint distribution through a factorization justi ed by statements of conditional independence which are encoded by a graph structure. Context-speci c independence is a natural generalization of conditional independence that only holds in a certain context, speci ed by the conditioning variables. We introduce context-speci c generalizations of both Bayesian networks and Markov networks by including statements of context-specific independence which can be encoded as a part of the model structures. For the purpose of learning context-speci c model structures from data, we derive score functions, based on results from Bayesian statistics, by which the plausibility of a structure is assessed. To identify high-scoring structures, we construct stochastic and deterministic search algorithms designed to exploit the structural decomposition of our score functions. Numerical experiments on synthetic and real-world data show that the increased exibility of context-specific structures can more accurately emulate the dependence structure among the variables and thereby improve the predictive accuracy of the models.

Spatial pattern detection modeling of thrips (Thrips tabaci) on onion fields

Onion (Allium cepa) is one of the most cultivated and consumed vegetables in Brazil and its importance is due to the large laborforce involved. One of the main pests that affect this crop is the Onion Thrips (Thrips tabaci), but the spatial distribution of this insect, although important, has not been considered in crop management recommendations, experimental planning or sampling procedures. Our purpose here is to consider statistical tools to detect and model spatial patterns of the occurrence of the onion thrips. In order to characterize the spatial distribution pattern of the Onion Thrips a survey was carried out to record the number of insects in each development phase on onion plant leaves, on different dates and sample locations, in four rural properties with neighboring farms under different infestation levels and planting methods. The Mantel randomization test proved to be a useful tool to test for spatial correlation which, when detected, was described by a mixed spatial Poisson model with a geostatistical random component and parameters allowing for a characterization of the spatial pattern, as well as the production of prediction maps of susceptibility to levels of infestation throughout the area.

Modeling associations between genetic markers using Bayesian networks

Motivation: Understanding the patterns of association between polymorphisms at different loci in a population ( linkage disequilibrium, LD) is of fundamental importance in various genetic studies. Many coefficients were proposed for measuring the degree of LD, but they provide only a static view of the current LD structure. Generative models (GMs) were proposed to go beyond these measures, giving not only a description of the actual LD structure but also a tool to help understanding the process that generated such structure. GMs based in coalescent theory have been the most appealing because they link LD to evolutionary factors. Nevertheless, the inference and parameter estimation of such models is still computationally challenging. Results: We present a more practical method to build GM that describe LD. The method is based on learning weighted Bayesian network structures from haplotype data, extracting equivalence structure classes and using them to model LD. The results obtained in public data from the HapMap database showed that the method is a promising tool for modeling LD. The associations represented by the learned models are correlated with the traditional measure of LD D`. The method was able to represent LD blocks found by standard tools. The granularity of the association blocks and the readability of the models can be controlled in the method. The results suggest that the causality information gained by our method can be useful to tell about the conservability of the genetic markers and to guide the selection of subset of representative markers.

Modeling extended Petri nets compatible with GHENeSys IEC61131 for industrial automation

Petri net (PN) modeling is one of the most used formal methods in the automation applications field, together with programmable logic controllers (PLCs). Therefore, the creation of a modeling methodology for PNs compatible with the IEC61131 standard is a necessity of automation specialists. Different works dealing with this subject have been carried out; they are presented in the first part of this paper [Frey (2000a, 2000b); Peng and Zhou (IEEE Trans Syst Man Cybern, Part C Appl Rev 34(4):523-531, 2004); Uzam and Jones (Int J Adv Manuf Technol 14(10):716-728, 1998)], but they do not present a completely compatible methodology with this standard. At the same time, they do not maintain the simplicity required for such applications, nor the use of all-graphical and all-mathematical ordinary Petri net (OPN) tools to facilitate model verification and validation. The proposal presented here completes these requirements. Educational applications at the USP and UEA (Brazil) and the UO (Cuba), as well as industrial applications in Brazil and Cuba, have already been carried out with good results.