Sammanfattning Sverige är ett kulturellt heterogent samhälle, och på skolorna runt om i Sverige finns därför en stor språklig variation. Det betyder att det ställs allt högre krav på lärare att hitta fungerande strategier för att stärka och stötta flerspråkiga elevers språkutveckling i relation till att det föreligger en resultat-klyfta mellan olika elevgrupper. Syfte Syftet med denna undersökning är att synliggöra hur grundskollärare förhåller sig till flerspråkiga elever och deras språkliga resurser samt undersöka vilka undervisningsmetoder som används i svenskundervisningen för att stötta och främja andraspråkselevers språkliga utveckling i svenskämnet. Frågeställningarna är; Hur beskriver lärare sitt förhållningssätt till flerspråkiga elever och deras språkliga resurser? samt Vilka undervisningsmetoder används för att stötta flerspråkiga elever i svenskundervisningen? Metod Studien utgår från en kvalitativ metod i form av intervjuer med sju stycken verksamma lärare. Denna undersökning avgränsades till att bara inkludera lågstadielärare. Studiens utgångspunkt är det sociokulturella perspektivet samt Jim Cummins teorier om andraspråksinlärning. Resultat Lärarna i denna studie har en positiv syn på flerspråkighet och betonar vikten av ett medvetet förhållningssätt och tryggt klassrumsklimat vid arbete med andraspråkselever. Vidare framgår det i resultatet att lärarna ger utrymme för eleverna att använda sina förstaspråk i klassrummet och att lärarna jobbar för att andraspråkselever ska utveckla sina kulturella och språkliga identiteter. Det framkommer även att lärarna behöver förhålla sig till flera svårigheter som är relaterade till att undervisa flerspråkiga elever, men att lärarna trots detta har en positiv syn på flerspråkighet. Vidare framgår det att lärarna använder flera olika metoder och strategier för att stötta andraspråkseleverna i deras språkliga utveckling. Samtliga lärare använder visuella stöd och digitala verktyg för att tydliggöra undervisningen, samt kooperativt lärande och mycket stöttning. Ytterligare en resurs som merparten av lärarna använder är studiehand-ledare. Det framkommer även att lärarna inte inkorporerar elevernas förstaspråk i den utsträckning som forskningen menar är lönsamt för flerspråkiga elever. Endast två lärare i denna studie använder en flerspråkig undervisningsstrategi, translanguaging. Studiens slutsats är att det finns en okunskap och osäkerhet bland verksamma lågstadielärare när det kommer till flerspråkiga undervisningsstrategier och att lärarna behöver fördjupade kunskaper om flerspråkig undervisning för att implementera det.

Excerpt: Yanan Song and Stephen Andrews address the topic of teachers’ use of the first language (L1) in the classroom by describing a case study of four tertiary-level teachers in China who share an L1 with their students.

No description available.

Studying molecular mechanisms underlying adipocyte differentiation is imperative to understanding adipocyte function and its role in obesity. However, the majority of research exploring adipogenesis is conducted with cell lines cultured directly on tissue culture plastic. Culturing cells on plastic may result in altered proliferation and differentiation, and subsequent change in pharmacological response. The extracellular matrix (ECM) plays a critical role in adipocyte development and survival. It is suggested that cells in vitro express high levels of ECM proteins to compensate for lack of an ECM. Differentiating preadipocytes on a substrate representative of the mature adipocyte extracellular environment may provide a more physiological response to drugs and environmental chemicals. The purpose of this study was to investigate the impact of Matrigel on 3T3-L1 cell growth, differentiation, lipid accumulation and responsiveness to Rosiglitazone. Matrigel decreased 3T3-L1 cell proliferation, enhanced lipid accumulation, and increased expression of adipogenic and lipogenic markers, including PPARγ, C/EBPα, SREBP1c, FAS, LPL, FABP4 and PLIN1. This was accompanied by a decrease in gene expression of ECM proteins, including fibronectin, collagen 1, collagen 3, collagen 4, laminin and collagen 6 in 3T3-L1 cells on Matrigel. Finally, Matrigel enhanced the response of 3T3-L1 cells to Rosiglitazone, which is a known PPARγ agonist and significantly increases lipid accumulation in 3T3-L1 cells. Our results suggest that enhanced lipid accumulation in 3T3-L1 cells on Matrigel is associated with decreased expression of ECM genes. Future studies require investigation of the cell-to-ECM interaction to confirm these findings. This study proposes that the nature of the ECM for cultured adipocytes alters temporal lipid accumulation patterns and response to various drugs as compared to 3T3-L1 cells grown on tissue culture plastic. / Thesis / Master of Science (MSc)

EFL/ESL teachers in Sweden have traditionally favored using only English in the classroom, however, research demonstrates they incorporate Swedish for various purposes. Moreover, teachers perceive the guidelines for the use of the target language and learners’ first language outlined in the curriculum for English as vague, experience feelings of guilt when resorting to Swedish in the English classroom and lack familiarity with the translanguaging theory. These issues could affect the requirement of evidence-based teaching and have implications for the quality of EFL/ESL education in Sweden. Through interviews and observations, this study aims to explore this matter and examine the reasons why and for what functions some English teachers in Sweden claim they use learners’ L1 in their teaching, as well as their level of awareness regarding their language practices and the concept of translanguaging. The findings indicate that the teachers primarily employ learners’ L1 to aid in knowledge construction, classroom management, and interpersonal relations. Additionally, certain inconsistencies were found between their motivations and their actual practices, indicating a potential lack of awareness and understanding of the translanguaging theory. Finally, it is suggested there is a necessity for clearer directives in the Swedish curriculum for English and for enhancing teachers’ awareness when implementing translanguaging strategies, such as pedagogical translanguaging (PTL).

Gene network modeling is one of the major goals of systems biology research. Gene network modeling targets the middle layer of active biological systems that orchestrate the activities of genes and proteins. Gene network modeling can provide critical information to bridge the gap between causes and effects which is essential to explain the mechanisms underlying disease. Among the network construction tasks, the rewiring of relevant network structure plays critical roles in determining the behavior of diseases. To systematically characterize the selectively activated regulatory components and mechanisms, the modeling tools must be able to effectively distinguish significant rewiring from random background fluctuations. While differential dependency networks cannot be constructed by existing knowledge alone, effective incorporation of prior knowledge into data-driven approaches can improve the robustness and biological relevance of network inference. Existing studies on protein-protein interactions and biological pathways provide constantly accumulated rich domain knowledge. Though novel incorporation of biological prior knowledge into network learning algorithms can effectively leverage domain knowledge, biological prior knowledge is neither condition-specific nor error-free, only serving as an aggregated source of partially-validated evidence under diverse experimental conditions. Hence, direct incorporation of imperfect and non-specific prior knowledge in specific problems is prone to errors and theoretically problematic. To address this challenge, we propose a novel mathematical formulation that enables incorporation of prior knowledge into structural learning of biological networks as Gaussian graphical models, utilizing the strengths of both measurement data and prior knowledge. We propose a novel strategy to estimate and control the impact of unavoidable false positives in the prior knowledge that fully exploits the evidence from data while obtains "second opinion" by efficient consultations with prior knowledge. By proposing a significance assessment scheme to detect statistically significant rewiring of the learned differential dependency network, our method can assign edge-specific p-values and specify edge types to indicate one of six biological scenarios. The data-knowledge jointly inferred gene networks are relatively simple to interpret, yet still convey considerable biological information. Experiments on extensive simulation data and comparison with peer methods demonstrate the effectiveness of knowledge-fused differential dependency network in revealing the statistically significant rewiring in biological networks, leveraging data-driven evidence and existing biological knowledge, while remaining robust to the false positive edges in the prior knowledge. We also made significant efforts in disseminating the developed method tools to the research community. We developed an accompanying R package and Cytoscape plugin to provide both batch processing ability and user-friendly graphic interfaces. With the comprehensive software tools, we apply our method to several practically important biological problems to study how yeast response to stress, to find the origin of ovarian cancer, and to evaluate the drug treatment effectiveness and other broader biological questions. In the yeast stress response study our findings corroborated existing literatures. A network distance measurement is defined based on KDDN and provided novel hypothesis on the origin of high-grade serous ovarian cancer. KDDN is also used in a novel integrated study of network biology and imaging in evaluating drug treatment of brain tumor. Applications to many other problems also received promising biological results. / Ph. D.

Dès les premiers instants de l’infection, le VIH-1 provoque un dérèglement généralisé du système immunitaire en altérant les fonctions et en favorisant un phénotype pro-inflammatoire de plusieurs types de cellules de l’immunité innée et adaptative. Certains mécanismes associés au dysfonctionnement lors des étapes précoces de l’infection restent à élucider. Lors d’une infection par le tractus génital, les cellules myéloïdes produisent la calprotectine, un complexe protéique pro-inflammatoire. Le but de ce projet de recherche était d’élucider certains mécanismes, associés à la calprotectine, contribuant au dérèglement du système immunitaire. L’analyse de l’effet de la calprotectine sur l’infection de lymphocytes T CD4 a montré qu’elle provoque une augmentation de la production virale et de l’infectivité des virions. Lors de ces expériences, les préparations virales contenaient aussi des vésicules extracellulaires, dont des exosomes. Le gradient de vélocité à l’iodixanol et l’immunocapture ont permis de retirer les exosomes des préparations virales et de montrer qu’ils n’avaient pas d’effet sur l’infectivité des virions. Toutefois, l’analyse du contenu en microARN dans le gradient de vélocité, fait à partir du plasma de personnes infectées, a révélé que le miR-155 pourrait être l’agent causal de l’augmentation de la capacité infectieuse des virions. Nous avons alors émis l’hypothèse que la calprotectine augmentait l’infectivité des virions en régulant positivement l’expression du miR-155, menant à son enrichissement dans des vésicules extracellulaires (VE) différentes des exosomes. Dans ce projet de recherche, nous avons montré que, parmi cinq microARN pouvant influencer le cours de l’infection par le VIH-1, la calprotectine augmente préférentiellement l’expression du miR-155 dans les lymphocytes T CD4 infectées en activant la transcription du gène miR-155 en miR-155 primaire (pri-miR-155). Nous avons aussi observé que la surexpression du miR-155 par les cellules provoquait un enrichissement de ce microARN dans les VE, particulièrement dans les VE se retrouvant dans les fractions les plus denses du gradient de vélocité. Finalement, les VE enrichies en miR-155 augmentent la production de virions, possiblement par l’inhibition de l’expression de SOCS-1. Ce projet a permis de mettre en lumière un mécanisme par lequel la calprotectine exacerbe l’infection par le VIH-1 et contribue à déréguler le fonctionnement de cellules immunitaires.

The discovery of immune checkpoints and the subsequent development of pharmacological immune checkpoint inhibitors generated excitement and optimism that researchers had finally found the key to inducing complete remission of tumors. Inhibitors targeting the PD-1/PD-L1 signaling axis, though still facing limitations with widespread utility, have proven to be beneficial in multiple types of cancerous tissues. While some patients experience dramatic results, immune checkpoint inhibitors are not the panacea that was initially hoped. Why anti-PD-1/anti-PD-L1 therapies often come up short might be that the underlying regulators of PD-L1 expression or other molecular regulators of immune suppression are not being addressed by the pharmacological immunomodulators. One protein often active in the tumor microenvironment (TME) is involved in both these possibilities: the aryl hydrocarbon receptor (AhR). The AhR is a ligand-activated transcription factor that can suppress immune activation as well as drive cancer progression by regulating transcription of a myriad of different genes within both malignant and immune cells. While AhR activation often correlates with poor prognosis in a variety of cancers, how AhR activation affects functional characteristics within and between malignant cells and immune cells in the TME is not fully understood. This is especially true in lung cancer, one of the most common and deadliest malignancies, which develops in an organ with many tissue-resident immune cells and opportunities to encounter exogenous pathogens—and environmental AhR agonists. To this end, we sought to elucidate the interactions between the AhR and its immunomodulatory target genes, PD-L1 and IDO1, in malignant lung adenocarcinoma (LUAD) cells and to determine how these interactions influence immunosuppression in the greater TME. The following work describes how AhR activation in LUAD cells results in the upregulation of two immune checkpoints, PD-L1 and IDO, in the malignant cell. We showed that AhR expression in CMT167 LUAD cells led to rapid tumor growth following subcutaneous or orthotopic transplantation whereas AhR deletion resulted in slower proliferation and up to complete rejection of tumor cells. AhR knockout also resulted in greater infiltration of CD3+ T cells. T cells from CMT167AhR-KO tumors were characterized by expression of granzyme B and upregulation of genes involved in activated and cytotoxic immune response, while T cells from CMT167WT tumors expressed higher levels of genes associated with T cell exhaustion and non-Th1 phenotypes. In addition, we developed a novel orthotopic model that will allow us to better understand how immune checkpoints from tumor cells modulate the lung TME by tracing the interactions of both myeloid and lymphoid immune cells with malignant cells. Further, in human LUAD cells, we implicated the AhR in the control of a long non-coding RNA shown to promote PD-L1 expression in the presence of IFNγ. Together, these results implicate AhR as a regulator of immune suppression in LUAD through multiple mechanisms and support it as a therapeutic target in the quest to improve immunotherapies.

Il existe des situations de modélisation statistique pour lesquelles le problème classique de classification non supervisée (c'est-à-dire sans information a priori sur la nature ou le nombre de classes à constituer) se double d'un problème d'identification des variables réellement pertinentes pour déterminer la classification. Cette problématique est d'autant plus essentielle que les données dites de grande dimension, comportant bien plus de variables que d'observations, se multiplient ces dernières années : données d'expression de gènes, classification de courbes... Nous proposons une procédure de sélection de variables pour la classification non supervisée adaptée aux problèmes de grande dimension. Nous envisageons une approche par modèles de mélange gaussien, ce qui nous permet de reformuler le problème de sélection des variables et du choix du nombre de classes en un problème global de sélection de modèle. Nous exploitons les propriétés de sélection de variables de la régularisation l1 pour construire efficacement, à partir des données, une collection de modèles qui reste de taille raisonnable même en grande dimension. Nous nous démarquons des procédures classiques de sélection de variables par régularisation l1 en ce qui concerne l'estimation des paramètres : dans chaque modèle, au lieu de considérer l'estimateur Lasso, nous calculons l'estimateur du maximum de vraisemblance. Ensuite, nous sélectionnons l'un des ces estimateurs du maximum de vraisemblance par un critère pénalisé non asymptotique basé sur l'heuristique de pente introduite par Birgé et Massart. D'un point de vue théorique, nous établissons un théorème de sélection de modèle pour l'estimation d'une densité par maximum de vraisemblance pour une collection aléatoire de modèles. Nous l'appliquons dans notre contexte pour trouver une forme de pénalité minimale pour notre critère pénalisé. D'un point de vue pratique, des simulations sont effectuées pour valider notre procédure, en particulier dans le cadre de la classification non supervisée de courbes. L'idée clé de notre procédure est de n'utiliser la régularisation l1 que pour constituer une collection restreinte de modèles et non pas aussi pour estimer les paramètres des modèles. Cette étape d'estimation est réalisée par maximum de vraisemblance. Cette procédure hybride nous est inspirée par une étude théorique menée dans une première partie dans laquelle nous établissons des inégalités oracle l1 pour le Lasso dans les cadres de régression gaussienne et de mélange de régressions gaussiennes, qui se démarquent des inégalités oracle l0 traditionnellement établies par leur absence totale d'hypothèse. / This thesis deals with variable selection for clustering. This problem has become all the more challenging since the recent increase in high-dimensional data where the number of variables can largely exceeds the number of observations (DNA analysis, functional data clustering...). We propose a variable selection procedure for clustering suited to high-dimensional contexts. We consider clustering based on finite Gaussian mixture models in order to recast both the variable selection and the choice of the number of clusters into a global model selection problem. We use the variable selection property of l1-regularization to build a data-driven model collection in a efficient way. Our procedure differs from classical procedures using l1-regularization as regards the estimation of the mixture parameters: in each model of the collection, rather than considering the Lasso estimator, we calculate the maximum likelihood estimator. Then, we select one of these maximum likelihood estimators by a non-asymptotic penalized criterion. From a theoretical viewpoint, we establish a model selection theorem for maximum likelihood estimators in a density estimation framework with a random model collection. We apply it in our context to determine a convenient penalty shape for our criterion. From a practical viewpoint, we carry out simulations to validate our procedure, for instance in the functional data clustering framework. The basic idea of our procedure, which consists in variable selection by l1-regularization but estimation by maximum likelihood estimators, comes from theoretical results we establish in the first part of this thesis: we provide l1-oracle inequalities for the Lasso in the regression framework, which are valid with no assumption at all contrary to the usual l0-oracle inequalities in the literature, thus suggesting a gap between l1-regularization and l0-regularization.

Hauptthema der Dissertation ist die Analysis eines nichtlinearen, gekoppelten Systems partieller Differentialgleichungen (PDG), das in der Modellierung der Kristallzüchtung aus der Schmelze mit Magnetfeldern vorkommt. Die zu beschreibenden Phenomäne sind einerseits der im elektromagnetisch geheizten Schmelzofen erfolgende Wärmetransport (Wärmeleitung, -konvektion und -strahlung), und andererseits die Bewegung der Halbleiterschmelze unter dem Einfluss der thermischen Konvektion und der angewendeten elektromagnetischen Kräfte. Das Modell besteht aus den Navier-Stokeschen Gleichungen für eine inkompressible Newtonsche Flüssigkeit, aus der Wärmeleitungsgleichung und aus der elektrotechnischen Näherung des Maxwellschen Systems. Wir erörtern die schwache Formulierung dieses PDG Systems, und wir stellen ein Anfang-Randwertproblem auf, das die Komplexität der Anwendung widerspiegelt. Die Hauptfrage unserer Untersuchung ist die Wohlgestelltheit dieses Problems, sowohl im stationären als auch im zeitabhängigen Fall. Wir zeigen die Existenz schwacher Lösungen in geometrischen Situationen, in welchen unstetige Materialeigenschaften und nichtglatte Trennfläche auftreten dürfen, und für allgemeine Daten. In der Lösung zum zeitabhängigen Problem tritt ein Defektmaß auf, das ausser der Flüssigkeit im Rand der elektrisch leitenden Materialien konzentriert bleibt. Da eine globale Abschätzung der im Strahlungshohlraum ausgestrahlten Wärme auch fehlt, rührt ein Teil dieses Defektmaßes von der nichtlokalen Strahlung her. Die Eindeutigkeit der schwachen Lösung erhalten wir nur unter verstärkten Annahmen: die Kleinheit der gegebenen elektrischen Leistung im stationären Fall, und die Regularität der Lösung im zeitabhängigen Fall. Regularitätseigenschaften wie die Beschränktheit der Temperatur werden, wenn auch nur in vereinfachten Situationen, hergeleitet: glatte Materialtrennfläche und Temperaturunabhängige Koeffiziente im Fall einer stationären Analysis, und entkoppeltes, zeitharmonisches Maxwell für das transiente Problem. / The present PhD thesis is devoted to the analysis of a coupled system of nonlinear partial differential equations (PDE), that arises in the modeling of crystal growth from the melt in magnetic fields. The phenomena described by the model are mainly the heat-transfer processes (by conduction, convection and radiation) taking place in a high-temperatures furnace heated electromagnetically, and the motion of a semiconducting melted material subject to buoyancy and applied electromagnetic forces. The model consists of the Navier-Stokes equations for a newtonian incompressible liquid, coupled to the heat equation and the low-frequency approximation of Maxwell''s equations. We propose a mathematical setting for this PDE system, we derive its weak formulation, and we formulate an (initial) boundary value problem that in the mean reflects the complexity of the real-life application. The well-posedness of this (initial) boundary value problem is the mainmatter of the investigation. We prove the existence of weak solutions allowing for general geometrical situations (discontinuous coefficients, nonsmooth material interfaces) and data, the most important requirement being only that the injected electrical power remains finite. For the time-dependent problem, a defect measure appears in the solution, which apart from the fluid remains concentrated in the boundary of the electrical conductors. In the absence of a global estimate on the radiation emitted in the cavity, a part of the defect measure is due to the nonlocal radiation effects. The uniqueness of the weak solution is obtained only under reinforced assumptions: smallness of the input power in the stationary case, and regularity of the solution in the time-dependent case. Regularity properties, such as the boundedness of temperature are also derived, but only in simplified settings: smooth interfaces and temperature-independent coefficients in the case of a stationary analysis, and, additionally for the transient problem, decoupled time-harmonic Maxwell.