Size-balanced community reorganization in response to nutrients and warming

It is widely accepted that global warming will adversely affect ecological communities. As ecosystems are simultaneously exposed to other anthropogenic influences, it is important to address the effects of climate change in the context of many stressors. Nutrient enrichment might offset some of the energy demands that warming can exert on organisms by stimulating growth at the base of the food web. It is important to know whether indirect effects of warming will be as ecologically significant as direct physiological effects. Declining body size is increasingly viewed as a universal response to warming, with the potential to alter trophic interactions. To address these issues, we used an outdoor array of marine mesocosms to examine the impacts of warming, nutrient enrichment and altered top-predator body size on a community comprised of the predator (shore crab Carcinus maenas), various grazing detritivores (amphipods) and algal resources. Warming increased mortality rates of crabs, but had no effect on their moulting rates. Nutrient enrichment and warming had near diametrically opposed effects on the assemblage, confirming that the ecological effects of these two stressors can cancel each other out. This suggests that nutrient-enriched systems might act as an energy refuge to populations of species under metabolic constraints due to warming. While there was a strong difference in assemblages between mesocosms containing crabs compared to mesocosms without crabs, decreasing crab size had no detectable effect on the amphipod or algal assemblages. This suggests that in allometrically balanced communities, the expected long-term effect of warming (declining body size) is not of similar ecological consequence to the direct physiological effects of warming, at least not over the six week duration of the experiment described here. More research is needed to determine the long-term effects of declining body size on the bioenergetic balance of natural communities.

Novel monoclonal antibody and peptide binders for Mycobacterium avium subsp. paratuberculosis and their application for magnetic separation

The generation of novel Mycobacterium avium subsp. paratuberculosis (MAP)-specific monoclonal antibodies and phage-display derived peptide binders, along with their application for the magnetic separation (MS) of MAP cells, is described. Our aim was to achieve even greater MAP capture capability than is possible with peptide-mediated magnetic separation (PMS) using a 50:50 mix of biotinylated-aMp3 and biotinylated-aMptD peptide-coated beads. Gamma-irradiated whole MAP cells and ethanol extracted antigens (EEA) from these cells were used to elicit an immune response and as phage-display biopanning targets. A range of novel binders was obtained and coated onto paramagnetic beads, both individually and in various combinations, for MS evaluation. IS900 PCR was employed after MS to provide quick results. Capture sensitivity was assessed using a range of MAP concentrations after which the most promising beads were tested for their specificity for MAP, by performing MS followed by culture using 10 other Mycobacterium species. Magnetic beads coated with the biotinylated EEA402 peptide demonstrated a greater level of MAP capture than the current PMS method, even when low numbers of MAP (<10 cfu/ml) were present; however these beads also captured a range of other mycobacteria and so lacked capture specificity. Magnetic beads coated with monoclonal antibodies 6G11 and 15D10 (used as a 50:50 mix or as dually coated beads) also demonstrated improved MAP capture relative to the current PMS method, but with little cross-reactivity to other Mycobacterium spp. Therefore, two new MS protocols are suggested, the application of which would be dependent upon the required endpoint. Biotinylated EEA402-coated beads could potentially be used with a MAP-specific PCR to ensure detection specificity, while beads coated with 6G11 and 15D10 monoclonal antibodies could be used with culture or the phage amplification assay.

Conversations with the Future: Creativity, Aesthetics and Industry Application in Anthropological Practice

This article presents a series of conversations with anthropologists working in collaborative, interdisciplinary settings and projects. It examines the changing role and place of anthropology on the island of Ireland, particular for early career anthropologists. With anthropologists now working in settings as diverse as business schools, health, music, documentary making and industry (to name but a few), early career researchers are now dealing with new challenges. In this piece, we map out a number of these debates and show how a multiplicity of anthropologies and practices are emerging on the island.

Optimal alarms systems and its application to financial time series

This thesis focuses on the application of optimal alarm systems to non linear time series models. The most common classes of models in the analysis of real-valued and integer-valued time series are described. The construction of optimal alarm systems is covered and its applications explored. Considering models with conditional heteroscedasticity, particular attention is given to the Fractionally Integrated Asymmetric Power ARCH, FIAPARCH(p; d; q) model and an optimal alarm system is implemented, following both classical and Bayesian methodologies. Taking into consideration the particular characteristics of the APARCH(p; q) representation for financial time series, the introduction of a possible counterpart for modelling time series of counts is proposed: the INteger-valued Asymmetric Power ARCH, INAPARCH(p; q). The probabilistic properties of the INAPARCH(1; 1) model are comprehensively studied, the conditional maximum likelihood (ML) estimation method is applied and the asymptotic properties of the conditional ML estimator are obtained. The final part of the work consists on the implementation of an optimal alarm system to the INAPARCH(1; 1) model. An application is presented to real data series.

Monte Carlo studies of phase transitions and adsorption : application to n-6 Lennard-Jones, C60 and zeolite systems

The solid-fluid transition properties of the n - 6 Lennard-Jones system are studied by means of extensive free energy calculations. Different values of the parameter n which regulates the steepness of the short-range repulsive interaction are investigated. Furthermore, the free energies of the n < 12 systems are calculated using the n = 12 system as a reference. The method relies on a generalization of the multiple histogram method that combines independent canonical ensemble simulations performed with different Hamiltonians and computes the free energy difference between them. The phase behavior of the fullerene C60 solid is studied by performing NPT simulations using atomistic models which treat each carbon in the molecule as a separate interaction site with additional bond charges. In particular, the transition from an orientationally frozen phase at low temperatures to one where the molecules are freely rotating at higher temperatures is studied as a function of applied pressure. The adsorption of molecular hydrogen in the zeolite NaA is investigated by means of grand-canonical Monte Carlo, in a wide range of temperatures and imposed gas pressures, and results are compared with available experimental data. A potential model is used that comprises three main interactions: van der Waals, Coulomb and induced polarization by the permanent electric field in the zeolite.

Nanoporous carbons from sisal residues and their application in hybrid TiO2/carbon photocatalysts for the removal and degradation of phenol in solution

Tese de doutoramento, Química (Química Tecnológica), Universidade de Lisboa, Faculdade de Ciências, 2016

Multilayer Perceptron Neural Networks Training Through Charged System Search and its Application for Non-Technical Losses Detection

The non-technical loss is not a problem with trivial solution or regional character and its minimization represents the guarantee of investments in product quality and maintenance of power systems, introduced by a competitive environment after the period of privatization in the national scene. In this paper, we show how to improve the training phase of a neural network-based classifier using a recently proposed meta-heuristic technique called Charged System Search, which is based on the interactions between electrically charged particles. The experiments were carried out in the context of non-technical loss in power distribution systems in a dataset obtained from a Brazilian electrical power company, and have demonstrated the robustness of the proposed technique against with several others natureinspired optimization techniques for training neural networks. Thus, it is possible to improve some applications on Smart Grids.

New molecularly-imprinted polymer for carnitine and its application as ionophore in potentiometric selective membranes

Carnitine (CRT) is a biological metabolite found in urine that contributes in assessingseveral disease conditions, including cancer. Novel quick screening procedures for CRT are therefore fundamental. This work proposes a novel potentiometric device where molecularly imprinted polymers (MIPs) were used as ionophores. The host-tailored sites were imprinted on a polymeric network assembled by radical polymerization of methacrylic acid (MAA) and trimethylpropane trimethacrylate (TRIM). Non-imprinted polymers (NIPs) were produced as control by removing the template from the reaction media. The selective membrane was prepared by dispersing MIP or NIP particles in plasticizer and poly(vinyl chloride), PVC, and casting this mixture over a solid contact support made of graphite. The composition of the selective membrane was investigated with regard to kind/amount of sensory material (MIP or NIP), and the need for a lipophilic additive. Overall, MIP sensors with additive exhibited the best performance, with near-Nernstian response down to ~ 1 × 10− 4 mol L− 1, at pH 5, and a detection limitof ~ 8 × 10− 5 mol L− 1. Suitable selectivity was found for all membranes, assessed by the matched potential method against some of the most common species in urine (urea, sodium, creatinine, sulfate, fructose and hemoglobin). CRT selective membranes including MIP materials were applied successfully to the potentiometric determination of CRT in urine samples.

Solid contact PVC membrane electrodes based on neutral or charged carriers for the selective reading of anionic sulfamethoxazole and their application to the analysis of aquaculture water

Sulfamethoxazole (SMX) is among the antibiotics employed in aquaculture for prophylactic and therapeutic reasons. Environmental and food spread may be prevented by controlling its levels in several stages of fish farming. The present work proposes for this purpose new SMX selective electrodes for the potentiometric determination of this sulphonamide in water. The selective membranes were made of polyvinyl chloride (PVC) with tetraphenylporphyrin manganese (III) chloride or cyclodextrin-based acting as ionophores. 2-nitrophenyl octyl ether was employed as plasticizer and tetraoctylammonium, dimethyldioctadecylammonium bromide or potassium tetrakis (4-chlorophenyl) borate was used as anionic or cationic additive. The best analytical performance was reported for ISEs of tetraphenylporphyrin manganese (III) chloride with 50% mol of potassium tetrakis (4-chlorophenyl) borate compared to ionophore. Nersntian behaviour was observed from 4.0 × 10−5 to 1.0 × 10−2 mol/L (10.0 to 2500 µg/mL), and the limit of detection was 1.2 × 10−5 mol/L (3.0 µg/mL). In general, the electrodes displayed steady potentials in the pH range of 6 to 9. Emf equilibrium was reached before 15 s in all concentration levels. The electrodes revealed good discriminating ability in environmental samples. The analytical application to contaminated waters showed recoveries from 96 to 106%.

Deconstructing Evidence-Based Practice and Its Application in Children to with Mental Health Disorders

This is a mixed methodology study that uses an autoethnographic approach to combine both an autobiography and a survey of practitioners who work in children’s mental health. It is largely about the implementation of Evidence-Based Practice (EBP), and the questions, concerns, experiences that I have had, and compared them with those of my fellow practitioners. In addition, it is about my journey as a mental health professional, and how I have come to recognize that in order to achieve the goals I wanted to achieve, I needed to return to university to pursue a Master’s degree. Within the research, I identify and discuss different definitions of EBP and identify several themes. I deconstruct the implementation of EBPs through the lens of Foucault and his notions of governmentality. I offer policy and practice recommendations to improve the implementation of EBP and the services received by children facing mental health issues.

Design of Resistivity Instrumentation for a He3 Cryostat and its Application to the Charge Density Wave Superconductor CuxTiSe2

Fermi patches in quasi-two dimensional charge density waves (CDW) have not described the connection to superconductivity (SC) according to theory adequately at this point in time. The connection between CDW and SC in the quasi-two dimensional material CuxTiSe2 is an interesting one which might reveal mechanisms in unconventional superconductors. A previous Brock graduate student grew crystals of CuxTiSe2. The precise doping of the samples was not known. In order to determine the doping parameter x in CuxTiSe2, a sensitive resistivity measurement system was necessary. A new resistivity measurement system was designed and implemented utilizing an Infrared Labs HDL-10 He3 cryostat. By comparing with data from the literature, doping of two samples was investigated using the new measurement system and a Quantum Design Magnetic Property Measurement System (MPMS). Methods for determining the doping revealed that the old resistivity system would not be able to determine the CDW transition temperature of highly doped samples or doping for elongated samples due to electronic noise. Doping in one sample was found to be between x=0.06 and x=0.065. Values of doping in the second sample had a discrepancy but could be explained by incorrect sample orientation.

Diastereoselective Synthesis of Planar Chiral N-Substituted Ferrocenes Derived from Epimeric Imidazolones and their Application to Asymmetric Hydrogenation of Quinolines

This thesis describes the synthesis and use of an N-substituted ferrocene bearing a proline-derived chiral directing group and diastereoselective lithiation-electrophile quench of the pro-Sp hydrogen of the ferrocene to give planar chiral products in >95:5 dr. The auxiliary group is found to be stable to lithium bases of types RLi and R2NLi giving the same diastereoselectivity. The anti- epimer of the previously mentioned syn auxiliary induces lithiation of pro Rp rather than pro Sp hydrogen in >95:5 dr. Upon electrophile quench and elimination, the enantiomer of the syn-derived planar chiral imidazolone is obtained. Hence, this method provides a practical way to prepare planar chiral enantiomers in this series without the use of a more expensive D-proline derived starting material. The syn and anti epimers have β, γ-stereogenic centers and the origin of stereoselectivity in lithiation appears to be driven by the conformational bias exerted by the β-silyloxy moiety in each chiral auxiliary. In the thesis, this conclusion is supported using insensitivity of lithiation selectivity to the bulkiness of the base, comparison of enantiomers, deuteration experiments, nOe difference studies and computational modeling of the ground states and lithiation transition states for both substrates. The products are then converted to ligand precursors to make iridium and rhodium complexes. Among them, one of the cationic iridium complex is found to be effective in the asymmetric hydrogenation of 2-substituted quinolines with enantioselectivities up to 80% at pressures as low as 5 atm.

Anatomo-functional magnetic resonance imaging of the spinal cord and its application to the characterization of spinal lesions in cats

Les lésions de la moelle épinière ont un impact significatif sur la qualité de la vie car elles peuvent induire des déficits moteurs (paralysie) et sensoriels. Ces déficits évoluent dans le temps à mesure que le système nerveux central se réorganise, en impliquant des mécanismes physiologiques et neurochimiques encore mal connus. L'ampleur de ces déficits ainsi que le processus de réhabilitation dépendent fortement des voies anatomiques qui ont été altérées dans la moelle épinière. Il est donc crucial de pouvoir attester l'intégrité de la matière blanche après une lésion spinale et évaluer quantitativement l'état fonctionnel des neurones spinaux. Un grand intérêt de l'imagerie par résonance magnétique (IRM) est qu'elle permet d'imager de façon non invasive les propriétés fonctionnelles et anatomiques du système nerveux central. Le premier objectif de ce projet de thèse a été de développer l'IRM de diffusion afin d'évaluer l'intégrité des axones de la matière blanche après une lésion médullaire. Le deuxième objectif a été d'évaluer dans quelle mesure l'IRM fonctionnelle permet de mesurer l'activité des neurones de la moelle épinière. Bien que largement appliquées au cerveau, l'IRM de diffusion et l'IRM fonctionnelle de la moelle épinière sont plus problématiques. Les difficultés associées à l'IRM de la moelle épinière relèvent de sa fine géométrie (environ 1 cm de diamètre chez l'humain), de la présence de mouvements d'origine physiologique (cardiaques et respiratoires) et de la présence d'artefacts de susceptibilité magnétique induits par les inhomogénéités de champ, notamment au niveau des disques intervertébraux et des poumons. L'objectif principal de cette thèse a donc été de développer des méthodes permettant de contourner ces difficultés. Ce développement a notamment reposé sur l'optimisation des paramètres d'acquisition d'images anatomiques, d'images pondérées en diffusion et de données fonctionnelles chez le chat et chez l'humain sur un IRM à 3 Tesla. En outre, diverses stratégies ont été étudiées afin de corriger les distorsions d'images induites par les artefacts de susceptibilité magnétique, et une étude a été menée sur la sensibilité et la spécificité de l'IRM fonctionnelle de la moelle épinière. Les résultats de ces études démontrent la faisabilité d'acquérir des images pondérées en diffusion de haute qualité, et d'évaluer l'intégrité de voies spinales spécifiques après lésion complète et partielle. De plus, l'activité des neurones spinaux a pu être détectée par IRM fonctionnelle chez des chats anesthésiés. Bien qu'encourageants, ces résultats mettent en lumière la nécessité de développer davantage ces nouvelles techniques. L'existence d'un outil de neuroimagerie fiable et robuste, capable de confirmer les paramètres cliniques, permettrait d'améliorer le diagnostic et le pronostic chez les patients atteints de lésions médullaires. Un des enjeux majeurs serait de suivre et de valider l'effet de diverses stratégies thérapeutiques. De telles outils représentent un espoir immense pour nombre de personnes souffrant de traumatismes et de maladies neurodégénératives telles que les lésions de la moelle épinière, les tumeurs spinales, la sclérose en plaques et la sclérose latérale amyotrophique.

A robust algorithm for segmenting fluorescence images and its application to single-molecule counting

La microscopie par fluorescence de cellules vivantes produit de grandes quantités de données. Ces données sont composées d’une grande diversité au niveau de la forme des objets d’intérêts et possèdent un ratio signaux/bruit très bas. Pour concevoir un pipeline d’algorithmes efficaces en traitement d’image de microscopie par fluorescence, il est important d’avoir une segmentation robuste et fiable étant donné que celle-ci constitue l’étape initiale du traitement d’image. Dans ce mémoire, je présente MinSeg, un algorithme de segmentation d’image de microscopie par fluorescence qui fait peu d’assomptions sur l’image et utilise des propriétés statistiques pour distinguer le signal par rapport au bruit. MinSeg ne fait pas d’assomption sur la taille ou la forme des objets contenus dans l’image. Par ce fait, il est donc applicable sur une grande variété d’images. Je présente aussi une suite d’algorithmes pour la quantification de petits complexes dans des expériences de microscopie par fluorescence de molécules simples utilisant l’algorithme de segmentation MinSeg. Cette suite d’algorithmes a été utilisée pour la quantification d’une protéine nommée CENP-A qui est une variante de l’histone H3. Par cette technique, nous avons trouvé que CENP-A est principalement présente sous forme de dimère.

Deep learning of representations and its application to computer vision

L’objectif de cette thèse par articles est de présenter modestement quelques étapes du parcours qui mènera (on espère) à une solution générale du problème de l’intelligence artificielle. Cette thèse contient quatre articles qui présentent chacun une différente nouvelle méthode d’inférence perceptive en utilisant l’apprentissage machine et, plus particulièrement, les réseaux neuronaux profonds. Chacun de ces documents met en évidence l’utilité de sa méthode proposée dans le cadre d’une tâche de vision par ordinateur. Ces méthodes sont applicables dans un contexte plus général, et dans certains cas elles on tété appliquées ailleurs, mais ceci ne sera pas abordé dans le contexte de cette de thèse. Dans le premier article, nous présentons deux nouveaux algorithmes d’inférence variationelle pour le modèle génératif d’images appelé codage parcimonieux “spike- and-slab” (CPSS). Ces méthodes d’inférence plus rapides nous permettent d’utiliser des modèles CPSS de tailles beaucoup plus grandes qu’auparavant. Nous démontrons qu’elles sont meilleures pour extraire des détecteur de caractéristiques quand très peu d’exemples étiquetés sont disponibles pour l’entraînement. Partant d’un modèle CPSS, nous construisons ensuite une architecture profonde, la machine de Boltzmann profonde partiellement dirigée (MBP-PD). Ce modèle a été conçu de manière à simplifier d’entraînement des machines de Boltzmann profondes qui nécessitent normalement une phase de pré-entraînement glouton pour chaque couche. Ce problème est réglé dans une certaine mesure, mais le coût d’inférence dans le nouveau modèle est relativement trop élevé pour permettre de l’utiliser de manière pratique. Dans le deuxième article, nous revenons au problème d’entraînement joint de machines de Boltzmann profondes. Cette fois, au lieu de changer de famille de modèles, nous introduisons un nouveau critère d’entraînement qui donne naissance aux machines de Boltzmann profondes à multiples prédictions (MBP-MP). Les MBP-MP sont entraînables en une seule étape et ont un meilleur taux de succès en classification que les MBP classiques. Elles s’entraînent aussi avec des méthodes variationelles standard au lieu de nécessiter un classificateur discriminant pour obtenir un bon taux de succès en classification. Par contre, un des inconvénients de tels modèles est leur incapacité de générer deséchantillons, mais ceci n’est pas trop grave puisque la performance de classification des machines de Boltzmann profondes n’est plus une priorité étant donné les dernières avancées en apprentissage supervisé. Malgré cela, les MBP-MP demeurent intéressantes parce qu’elles sont capable d’accomplir certaines tâches que des modèles purement supervisés ne peuvent pas faire, telles que celle de classifier des données incomplètes ou encore celle de combler intelligemment l’information manquante dans ces données incomplètes. Le travail présenté dans cette thèse s’est déroulé au milieu d’une période de transformations importantes du domaine de l’apprentissage à réseaux neuronaux profonds qui a été déclenchée par la découverte de l’algorithme de “dropout” par Geoffrey Hinton. Dropout rend possible un entraînement purement supervisé d’architectures de propagation unidirectionnel sans être exposé au danger de sur- entraînement. Le troisième article présenté dans cette thèse introduit une nouvelle fonction d’activation spécialement con ̧cue pour aller avec l’algorithme de Dropout. Cette fonction d’activation, appelée maxout, permet l’utilisation de aggrégation multi-canal dans un contexte d’apprentissage purement supervisé. Nous démontrons comment plusieurs tâches de reconnaissance d’objets sont mieux accomplies par l’utilisation de maxout. Pour terminer, sont présentons un vrai cas d’utilisation dans l’industrie pour la transcription d’adresses de maisons à plusieurs chiffres. En combinant maxout avec une nouvelle sorte de couche de sortie pour des réseaux neuronaux de convolution, nous démontrons qu’il est possible d’atteindre un taux de succès comparable à celui des humains sur un ensemble de données coriace constitué de photos prises par les voitures de Google. Ce système a été déployé avec succès chez Google pour lire environ cent million d’adresses de maisons.