Trophic structure in a seabird host-parasite food web: insights from stable isotope analyses

Ecological studies on food webs rarely include parasites, partly due to the complexity and dimensionality of host-parasite interaction networks. Multiple co-occurring parasites can show different feeding strategies and thus lead to complex and cryptic trophic relationships, which are often difficult to disentangle by traditional methods. We analyzed stable isotope ratios of C (13C/12C, δ13C) and N (15N/14N, δ15N) of host and ectoparasite tissues to investigate trophic structure in 4 co-occurring ectoparasites: three lice and one flea species, on two closely related and spatially segregated seabird hosts (Calonectris shearwaters). δ13C isotopic signatures confirmed feathers as the main food resource for the three lice species and blood for the flea species. All ectoparasite species showed a significant enrichment in δ15N relatively to the host tissue consumed (discrimination factors ranged from 2 to 5 depending on the species). Isotopic differences were consistent across multiple host-ectoparasite locations, despite of some geographic variability in baseline isotopic levels. Our findings illustrate the influence of both ectoparasite and host trophic ecology in the isotopic structuring of the Calonectris ectoparasite community. This study highlights the potential of stable isotope analyses in disentangling the nature and complexity of trophic relationships in symbiotic systems.

A nonparametric method for the measurement of size diversity with emphasis on data standardization

The most suitable method for estimation of size diversity is investigated. Size diversity is computed on the basis of the Shannon diversity expression adapted for continuous variables, such as size. It takes the form of an integral involving the probability density function (pdf) of the size of the individuals. Different approaches for the estimation of pdf are compared: parametric methods, assuming that data come from a determinate family of pdfs, and nonparametric methods, where pdf is estimated using some kind of local evaluation. Exponential, generalized Pareto, normal, and log-normal distributions have been used to generate simulated samples using estimated parameters from real samples. Nonparametric methods include discrete computation of data histograms based on size intervals and continuous kernel estimation of pdf. Kernel approach gives accurate estimation of size diversity, whilst parametric methods are only useful when the reference distribution have similar shape to the real one. Special attention is given for data standardization. The division of data by the sample geometric mean is proposedas the most suitable standardization method, which shows additional advantages: the same size diversity value is obtained when using original size or log-transformed data, and size measurements with different dimensionality (longitudes, areas, volumes or biomasses) may be immediately compared with the simple addition of ln k where kis the dimensionality (1, 2, or 3, respectively). Thus, the kernel estimation, after data standardization by division of sample geometric mean, arises as the most reliable and generalizable method of size diversity evaluation

Diferencias en la dimensionalidad electroencefalograma entre vigilia y sueño profundo

Differences in dimensionality of electroencephalogram during awake and deeper sleep stages. The nonlinear dynamical systems theory provides some tools for the analysis of electroencephalogram (EEG) at different sleep stages. Its use could allow the automatic monitoring of the states of the sleep and it would also contribute an explanatory level of the differences between stages. The goal of the present paper is to address this type of analysis, focusing on the most different stages. Estimations of dimensionality were compared when six subjects were awake and in a deep sleep stage. Greater dimensionality involves more complexity because the system receives more external influences. If this dimensionality is maximum, we can consider that the time series is a noisy one. A smaller dimensionality involves lower complexity because the system receives fewer inputs. We hypothesized that we would find greater dimensionality when subjects were awake than in a deep sleep stage. Results show a noisy time series during the awake stage, whereas in the sleep stage, dimensionality is smaller, confirming our hypothesis. This result is similar to the findings reached previously by other authors.

Estratégia supramolecular para a nanotecnologia

Brazilian science is evolving rapidly and steadly in the last 10 years, reaching the 15º place in the international ranking. Research in nanotechnology is following a similar way generating new scientific and technological knowledge in several frontiers but specially in the interfaces of two or more areas, where Chemistry is consolidating itself as a central science. In this context, the supramolecular approach is a very promissing one because it allows the build-up of a chemical inteligence using all the sistematized knowledge for the design and development of new nanomaterials and products. The great challenge of Chemistry is not decrease the dimensionality of the materials but instead find ways to increase the dimensionality and structural complexity keeping strict control on the interactions between the components, in order to generate materials with new properties and functionalities. Unfortunately, the current vigorous advancement of scientific research has not been followed by the transformation of such know-how into patents and produts. Therefore much efforts should be devoted to build a national science and technology program, joining all the segments of the society involved in the technological development (university, institutes of technological research, industry and government) in order to promote the furtherance of the Brazilian technological base. Only in this way it is possible to evolve to a technological society capable to transform the scientific knowledge into wealthy, thus sustaining the socioeconomic development of the country.

Measuring Brand Equity and Analyzing Its Relationship with Key Marketing Mix Elements

The aim of this paper is to analyze the effect of price and advertising on brand equity. The dimensionality of brand equity is thoroughly examined, and the effect price, price deals, perceived advertising spending and advertising appeal have on the dimensions of brand equity are analyzed using multiple regression analysis as well as other supporting analyses. Price and advertising are found to be of great importance to brand equity. Arguably the most influential finding is the strong positive effect low prices – an integral brand element – have on the case company brand equity, even though a negative effect was hypothesized based on prior research. The results also support separating advertising appeal from perceived advertising spending, as well as linking service quality as part of the overall perceived quality in the context of service-intensive firms.

Motivational Influences on Transfer: Dimensions and Boundary Conditions

The focus of this dissertation is the motivational influences on transfer in higher education and professional training contexts. To estimate these motivational influences, the dissertation includes seven individual studies that are structured in two parts. Part I, Dimensions, aims at identifying the dimensionality of motivation to transfer and its structural relations with training-related antecedents and outcomes. Part II, Boundary Conditions, aims at testing the predictive validity of motivation theories used in contemporary training research under different study conditions. Data in this dissertation was gathered from multi-item questionnaires, which were analyzed differently in Part I and Part II. Studies in Part I employed exploratory and confirmatory factor analysis, structural equation modeling, partial least squares (PLS) path modeling, and mediation analysis. Studies in Part II used artifact distribution meta-analysis, (nested) subgroup analysis, and weighted least squares (WLS) multiple regression. Results demonstrate that motivation to transfer can be conceptualized as a three-dimensional construct, including autonomous motivation to transfer, controlled motivation to transfer, and intention to transfer, given a theoretical framework informed by expectancy theory, self-determination theory, and the theory of planned behavior. Results also demonstrate that a range of boundary conditions moderates motivational influences on transfer. To test the predictive validity of expectancy theory, social cognitive theory, and the theory of goal orientations under different study settings, a total of 17 boundary conditions were meta-analyzed, including age; assessment criterion; assessment source; attendance policy; collaboration among trainees; computer support; instruction; instrument used to measure motivation; level of education; publication type; social training context; SS/SMC bias; study setting; survey modality; type of knowledge being trained; use of a control group; and work context. Together, the findings cumulated in this thesis support the basic premise that motivation is centrally important for transfer, but that motivational influences need to be understood from a more differentiated perspective than commonly found in the literature, in order to account for several dimensions and boundary conditions. The results of this dissertation across the seven individual studies are reflected in terms of their implications for theory development and their significance for training evaluation and the design of training environments. Limitations and directions to take in future research are discussed.

Feature selection using Fuzzy Entropy measures with Yu's Similarity measure

In this study, feature selection in classification based problems is highlighted. The role of feature selection methods is to select important features by discarding redundant and irrelevant features in the data set, we investigated this case by using fuzzy entropy measures. We developed fuzzy entropy based feature selection method using Yu's similarity and test this using similarity classifier. As the similarity classifier we used Yu's similarity, we tested our similarity on the real world data set which is dermatological data set. By performing feature selection based on fuzzy entropy measures before classification on our data set the empirical results were very promising, the highest classification accuracy of 98.83% was achieved when testing our similarity measure to the data set. The achieved results were then compared with some other results previously obtained using different similarity classifiers, the obtained results show better accuracy than the one achieved before. The used methods helped to reduce the dimensionality of the used data set, to speed up the computation time of a learning algorithm and therefore have simplified the classification task

Efficient Optimization Algorithms for Nonlinear Data Analysis

Identification of low-dimensional structures and main sources of variation from multivariate data are fundamental tasks in data analysis. Many methods aimed at these tasks involve solution of an optimization problem. Thus, the objective of this thesis is to develop computationally efficient and theoretically justified methods for solving such problems. Most of the thesis is based on a statistical model, where ridges of the density estimated from the data are considered as relevant features. Finding ridges, that are generalized maxima, necessitates development of advanced optimization methods. An efficient and convergent trust region Newton method for projecting a point onto a ridge of the underlying density is developed for this purpose. The method is utilized in a differential equation-based approach for tracing ridges and computing projection coordinates along them. The density estimation is done nonparametrically by using Gaussian kernels. This allows application of ridge-based methods with only mild assumptions on the underlying structure of the data. The statistical model and the ridge finding methods are adapted to two different applications. The first one is extraction of curvilinear structures from noisy data mixed with background clutter. The second one is a novel nonlinear generalization of principal component analysis (PCA) and its extension to time series data. The methods have a wide range of potential applications, where most of the earlier approaches are inadequate. Examples include identification of faults from seismic data and identification of filaments from cosmological data. Applicability of the nonlinear PCA to climate analysis and reconstruction of periodic patterns from noisy time series data are also demonstrated. Other contributions of the thesis include development of an efficient semidefinite optimization method for embedding graphs into the Euclidean space. The method produces structure-preserving embeddings that maximize interpoint distances. It is primarily developed for dimensionality reduction, but has also potential applications in graph theory and various areas of physics, chemistry and engineering. Asymptotic behaviour of ridges and maxima of Gaussian kernel densities is also investigated when the kernel bandwidth approaches infinity. The results are applied to the nonlinear PCA and to finding significant maxima of such densities, which is a typical problem in visual object tracking.

Dimensions and dynamics in integrin function

Integrins play crucial roles in cell adhesion, migration, and signaling by providing transmembrane links between the extracellular matrix and the cytoskeleton. Integrins cluster in macromolecular complexes to generate cell-matrix adhesions such as focal adhesions. In this mini-review, we compare certain integrin-based biological responses and signaling during cell interactions with standard 2D cell culture versus 3D matrices. Besides responding to the composition of the matrix, cells sense and react to physical properties that include three-dimensionality and rigidity. In routine cell culture, fibroblasts and mesenchymal cells appear to use focal adhesions as anchors. They then use intracellular actomyosin contractility and dynamic, directional integrin movements to stretch cell-surface fibronectin and to generate characteristic long fibrils of fibronectin in "fibrillar adhesions". Some cells in culture proceed to produce dense, three-dimensional matrices similar to in vivo matrix, as opposed to the flat, rigid, two-dimensional surfaces habitually used for cell culture. Cells within such more natural 3D matrices form a distinctive class of adhesion termed "3D-matrix adhesions". These 3D adhesions show distinctive morphology and molecular composition. Their formation is heavily dependent on interactions between integrin alpha5ß1 and fibronectin. Cells adhere much more rapidly to 3D matrices. They also show more rapid morphological changes, migration, and proliferation compared to most 2D matrices or 3D collagen gels. Particularly notable are low levels of tyrosine phosphorylation of focal adhesion kinase and moderate increases in activated mitogen-activated protein kinase. These findings underscore the importance of the dimensionality and dynamics of matrix substrates in cellular responses to the extracellular matrix.

Self concept, acculturation, and fashion orientation

The research begins with a discussion of the worldwide and the Canadian market. The research profiles the examination of the relationship between a person's self concept (as defined by Malhotra) and fashion orientation (as defined by Gutman and Mills), and to understand how these factors are influenced by acculturation, focusing in-depth on their managerial implications. To study these relationships; a random sample of 196 ChineseCanadian female university students living in Canada was given a survey based on Malhotra's self-concept scale, and the SLASIA acculturation scale. Based on multiple regression analysis, findings suggest that the adoption of language and social interaction dimensions of acculturation constructs have significant effects on the relationship between self concept and fashion orientation. This research contributes significantly to both marketing theory and practice. Theoretically, this research develops new insights on the dimensionality of fashion orientation, identifies various moderating effects of acculturation on the relationship of self concept and fashion orientation dimensions, and provides a framework to examine these effects, where results can be generalized across different culture. Practically, marketers can use available findings to improve their understanding of the fashion needs of Chinese-Canadian consumers, and target them based on these findings. The findings provide valuable implications for companies to formulate their fashion marketing strategies for enhance fashion orientation in terms of different dimensions, based on different levels of acculturation.

Comparison of classification ability of hyperball algorithms to neural network and k-nearest neighbour algorithms

The main focus of this thesis is to evaluate and compare Hyperbalilearning algorithm (HBL) to other learning algorithms. In this work HBL is compared to feed forward artificial neural networks using back propagation learning, K-nearest neighbor and 103 algorithms. In order to evaluate the similarity of these algorithms, we carried out three experiments using nine benchmark data sets from UCI machine learning repository. The first experiment compares HBL to other algorithms when sample size of dataset is changing. The second experiment compares HBL to other algorithms when dimensionality of data changes. The last experiment compares HBL to other algorithms according to the level of agreement to data target values. Our observations in general showed, considering classification accuracy as a measure, HBL is performing as good as most ANn variants. Additionally, we also deduced that HBL.:s classification accuracy outperforms 103's and K-nearest neighbour's for the selected data sets.

Crystal Growth and Characterization of Li0.9Mo6O17

Purple bronze Li0.9Mo6O17 has attracted researchers for its low dimensionality and corresponding properties. Although it has been studied for nearly two decades, there are still some unsolved puzzles with this unique material. Single crystals of Li0.9Mo6O17 were grown using the temperature gradient flux technique in this research. The crystal growth was optimized by experimenting different conditions and good quality crystals were obtained. X-ray diffraction results have confirmed the right phase of the crystals. Resistivity measurements and magnetic susceptibility measurements were carried out, and anomalous electronic behaviors were found. All of the samples showed the metal-insulator transition near 20K, followed by behavior that differs from sample to sample: either superconducting, metallic or insulating behavior was observed below 2K. Li0.9Mo6O17 was considered as a quasi-one-dimensional crystal and also a superconducting crystal, which implies a dimensional crossover may occur at the metal-insulator transition. A two-band scenario of the Luttinger liquid model was used to fit the resistivity data and excellent results were achieved, suggesting that the Luttinger theory is a very good candidate for the explanation of the anomalous behavior of Li0.9Mo6O17. In addition, the susceptibility measurements showed Curie paramagnetism and some temperature independent paramagnetism at low temperature. The absence of any anomalous magnetic feature near 20K where the resistivity upturn takes place, suggests that a charge density wave mechanism, which has been proposed by some researchers, is not responsible for the unique properties of Li0.9Mo6O17.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and deterministic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel metaheuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS metaheuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and determinis- tic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel meta–heuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS meta–heuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

Genes involved in the metabolism of fatty acids and risk for Crohn's disease in children: a candidate gene study

Contexte - La prévalence de la maladie de Crohn (MC), une maladie inflammatoire chronique du tube digestif, chez les enfants canadiens se situe parmi les plus élevées au monde. Les interactions entre les réponses immunes innées et acquises aux microbes de l'hôte pourraient être à la base de la transition de l’inflammation physiologique à une inflammation pathologique. Le leucotriène B4 (LTB4) est un modulateur clé de l'inflammation et a été associé à la MC. Nous avons postulé que les principaux gènes impliqués dans la voie métabolique du LTB4 pourrait conférer une susceptibilité accrue à l'apparition précoce de la MC. Dans cette étude, nous avons exploré les associations potentielles entre les variantes de l'ADN des gènes ALOX5 et CYP4F2 et la survenue précoce de la MC. Nous avons également examiné si les gènes sélectionnés montraient des effets parent-d'origine, influençaient les phénotypes cliniques de la MC et s'il existait des interactions gène-gène qui modifieraient la susceptibilité à développer la MC chez l’enfant. Méthodes – Dans le cadre d’une étude de cas-parents et de cas-témoins, des cas confirmés, leurs parents et des contrôles ont été recrutés à partir de trois cliniques de gastro-entérologie à travers le Canada. Les associations entre les polymorphismes de remplacement d'un nucléotide simple (SNP) dans les gènes CYP4F2 et ALOX5 ont été examinées. Les associations allélique et génotypiques ont été examinées à partir d’une analyse du génotype conditionnel à la parenté (CPG) pour le résultats cas-parents et à l’aide de table de contingence et de régression logistique pour les données de cas-contrôles. Les interactions gène-gène ont été explorées à l'aide de méthodes de réduction multi-factorielles de dimensionnalité (MDR). Résultats – L’étude de cas-parents a été menée sur 160 trios. L’analyse CPG pour 14 tag-SNP (10 dans la CYP4F2 et 4 dans le gène ALOX5) a révélé la présence d’associations alléliques ou génotypique significatives entre 3 tag-SNP dans le gène CYP4F2 (rs1272, p = 0,04, rs3093158, p = 0.00003, et rs3093145, p = 0,02). Aucune association avec les SNPs de ALOX5 n’a pu être démontrée. L’analyse de l’haplotype de CYP4F2 a montré d'importantes associations avec la MC (test omnibus p = 0,035). Deux haplotypes (GAGTTCGTAA, p = 0,05; GGCCTCGTCG, p = 0,001) montraient des signes d'association avec la MC. Aucun effet parent-d'origine n’a été observé. Les tentatives de réplication pour trois SNPs du gene CYP4F2 dans l'étude cas-témoins comportant 225 cas de MC et 330 contrôles suggèrent l’association dans un de ceux-ci (rs3093158, valeur non-corrigée de p du test unilatéral = 0,03 ; valeur corrigée de p = 0.09). La combinaison des ces deux études a révélé des interactions significatives entre les gènes CYP4F2, ALOX et NOD2. Nous n’avons pu mettre en évidence aucune interaction gène-sexe, de même qu’aucun gène associé aux phénotypes cliniques de la MC n’a pu être identifié. Conclusions - Notre étude suggère que la CYP4F2, un membre clé de la voie métabolique LTB4 est un gène candidat potentiel pour MC. Nous avons également pu mettre en évidence que les interactions entre les gènes de l'immunité adaptative (CYP4F2 et ALOX5) et les gènes de l'immunité innée (NOD2) modifient les risques de MC chez les enfants. D'autres études sur des cohortes plus importantes sont nécessaires pour confirmer ces conclusions.