The thesis work is conducted in the division of computer engineering at thedepartment of electrical engineering in Linköping University. During the thesiswork, a configurable Direct Memory Access (DMA) controller was designed andimplemented. The DMA controller runs at 200MHz under 65nm digital CMOS technology. The estimated gate count is 26595. The DMA controller has two address generators and can provide two clocksources. It can thus handle data read and write simultaneously. There are 16channels built in the DMA controller, the data width can be 16-bit, 32-bit and64-bit. The DMA controller supports 2D data access by configuring its intelligentlinking table. The DMA is designed for advanced DSP applications and it is notdedicated for cache which has a fixed priority.

L’objectif du projet est de comprendre l’effet du peptide vecteur (R/W)9 sur les cellules tumorales EF, modèles du sarcome d’Ewing. En effet, ce peptide a la capacité de remodeler le cytosquelette d’actine dans ces cellules, ainsi que de réduire leur motilité et leur aptitude à croître en indépendance d’ancrage (Delaroche D. JBC 2010).La première étape de ce travail a été la caractérisation in vitro de l’interaction directe avec l’actine par cross-linking chimique et spectrométrie de masse (Clavier S. EuProt 2014). Ensuite, pour avancer dans la compréhension de l’effet du peptide (R/W)9, deux approches ont été développées.La première approche, basée sur du photocross-linking in cellulo ou in lysat, vise à identifier des partenaires intracellulaires du peptide vecteur. Pour cela, nous avons mis au point et validé biologiquement une version photoactivable du peptide (R/W)9. Puis, avant de passer sur cellules entières, nous nous sommes assurés de la faisabilité de la réaction de photocross-linking in vitro sur un système d’interaction modèle que nous avons également utilisé pour développer un logiciel capable d’interpréter les spectres MS/MS d’espèces photocross-linkées (Xlink-Identifier, Collaboration Dr Du X). Des expériences de purification d’affinité ont également été menées en immobilisant le peptide (R/W)9 sur des billes de streptavidine ensuite mises en présence de lysat cellulaire. Les protéines capturées ont été identifiées par spectrométrie de masse haut-débit. La seconde approche est de la protéomique différentielle avec un marquage SILAC et a pour objectif de mettre en évidence l’influence du peptide vecteur sur l’expression des protéines. / The aim of the project is to understand the effect of (R/W)9 cell penetrating peptide (CPP) on EF tumoral cells, an Ewing sarcoma model cell line. Actually, this peptide is able to remodel the actin cytoskeleton of these cells, to decrease their motility as well as their ability to grow without anchorage (Delaroche D. JBC 2010).The first step of this work was to characterize the in vitro interaction with actin using chemical cross-linking and mass spectrometry (Clavier S. Euprot 2014). Then, in order to get a deeper understanding of (R/W)9 peptide effect, two approaches were developed. The goal of the first approach based on in cellulo or in lysate photocross-linking is to identify (R/W)9 CPP’s partners. To do this, we designed and biologically validated a photoactivable version of (R/W)9 peptide. Then, before starting to work with living cells, we checked the feasibility of in vitro photocross-linking on a model interacting system that we also used to develop a software able to interpret MS/MS spectra of photocross-linked species. (Xlink-Identifier, Collaboration with Dr. X. Du.). Affinity purification experiments were also performed by incubating streptavidin magnetic beads bearing (R/W)9 peptide with cell lysates. Captured proteins were identified using high-throughput mass spectrometry. The second approach is differential proteomic with SILAC labelling and aims at assessing the influence of (R/W)9 CPP on proteins expression.

L’étude du système verbal du tibétain littéraire présente un intérêt typologique à plus d’un titre. D’une part, elle permet d’observer l’évolution du système verbal, notamment les constructions avec un auxiliaire et les connecteurs verbaux, sur une période de plus de mille ans. Cette langue classique a l’avantage d’avoir quasiment préservé la même orthographe au cours de cette très longue période. J’ai choisi de me concentrer sur la période du quinzième siècle en choisissant pour corpus principal une œuvre très célèbre : la biographie de Milarépa. J’ai ensuite comparé le système verbal de cette période avec le tibétain littéraire contemporain. La principale raison qui a motivé cette étude, est l’émergence en tibétain moyen d’un système d’auxiliaires indiquant l’évidentialité, c'est-à-dire la grammaticalisation de la source épistémologique et de l’accès à l’information. Le tibétain est la seule langue littéraire d’Asie ayant une grande ancienneté qui ait développé un système verbal évidentiel complexe. Outre l’étude de la sémantique grammaticale, la deuxième motivation à l’origine de cette étude est la syntaxe des constructions avec un auxiliaire et des connecteurs du tibétain moyen et leur évolution en tibétain littéraire contemporain. Les données du tibétain littéraire et les analyses synchronique et diachronique sont susceptibles d’apporter une contribution à la typologie aussi bien en ce qui concerne l’évidentialité et les modalités épistémiques qu’en ce qui concerne la syntaxe des auxiliaires et des connecteurs. / The analysis of the literary Tibetan verb system is an object of typological interest for several reasons. Firstly, it allows us to look at the evolution of the verb system notably auxiliary verb constructions and clause linking during a period of more than a thousand years. Classical Tibetan has the advantage of having preserved almost the same orthography during this very long period of time. I decided to focus on the fifteenth century by selecting a well-known book as a main corpus: the Life of Milarepa. I then compared the verb system of this period with contemporary literary Tibetan. The main reason for this study is to better understand the emergence of an auxiliary verb system in middle Tibetan which marks evidentiality, that is to say the grammaticalization of the epistemological source and the access to information. Literary Tibetan is the only language in Asia with an ancient history which has developed a complex evidential verb system. Apart from the analysis of grammatical semantics, the second reason for this study is my interest in the syntax of auxiliary verb constructions and of clause linking in middle Tibetan and their evolution in contemporary Tibe! tan. My data on literary Tibetan and my synchronic and diachronic analysis may make a significant contribution to the typological studies of evidentiality and epistemic modality as well as of the syntax of auxiliary verb constructions and clause linking.

The goal of this master thesis is to design and implement a named entity recognition and linking algorithm. A part of this goal is to propose and create a knowledge base that will be used in the algorithm. Because of the limited amount of data for languages other than English, we want to be able to train our method on one language, and then transfer the learned parameters to other languages (that do not have enough training data). The thesis consists of description of available knowledge bases, existing methods and design and implementation of our own knowledge base and entity linking method. Our method achieves state of the art result on a few variants of the AIDA CoNLL-YAGO dataset. The method also obtains comparable results on a sample of Czech annotated data from the PDT dataset using the parameters trained on the English CoNLL dataset. Powered by TCPDF (www.tcpdf.org)

Le cadre industriel de ce projet est de trouver une alternative à la résine phénol-formaldéhyde (RFP) et à son durcisseur, utilisée actuellement en pneumatique pour améliorer les performances aussi bien des compositions de caoutchouc que celles des produits semi-finis. Un des aspects essentiels de la pneumatique est d’augmenter la rigidité à faible déformation des pneus sans augmenter l'hystérésis des élastomères durcis chargés de noir de carbone. Ainsi, l'objectif de ce travail de thèse est de proposer un système thermodurcissable alternatif aux résines RFP actuelles. Ce nouveau système doit être riche en carbone renouvelable et doit être chimiquement modifié au moyen de procédés écologiques. La lignine a été choisie comme source de carbone renouvelable. Une étude préliminaire a souligné les potentialités d'une classe de lignines époxy pour l'application visée. Notre approche consiste en la conception de lignines modifiées par des époxy par une méthode originale évitant l'utilisation de l'épichlorhydrine suivie de son durcissement par des agents de réticulation appropriés. Des tests préliminaires avec un composé modèle de lignine ont donné la gamme des composés époxy et les conditions à tester pour la réaction avec la lignine. Les expériences avec la lignine ont alors confirmé le potentiel de l’ester diglycidylique de l’acide 4, 5-époxytetrahydrophthalique, un composé époxy contenant à la fois un groupement cycloaliphatique et deux types de groupements époxy glycidyliques, comme alternative à l'épichlorohydrine pour la préparation de lignines modifiées. Dans les conditions que nous avons définies, la réaction donne des prépolymères de type lignine-époxy sous forme de poudre avec un niveau d’époxydation de 1,2 mol/kg déterminé par spectroscopie FTIR. La série de tests effectuée sur des mélanges de caoutchouc a montré que la nouvelle lignine époxy associée à la p-xylylènediamine peut être mélangée avec succès avec le caoutchouc naturel et peut atteindre les propriétés de notre mix de référence. / The industrial framework of this project is the substitution of phenol formaldehyde resin (RFP) and its methyl donor hardener currently used for improving the performances of rubber compositions of tires or semi-finished products for tires. A critical aspect is the need for an increase of rigidity at low deformation without enhancement of the hysteresis of the cured elastomers filled with carbon black. The specific aim of this PhD work was to propose an alternative thermosetting system to the current RFP resins. This new system should be rich in renewable carbon and chemically modified by environmentally friendly processes. Lignin was chosen as the source of the renewable carbon. A preliminary study emphasized the potentialities of a class of epoxy-modified lignins for the targeted application. The main approach was the design of epoxy modified lignin to be obtained by an original method avoiding the use of epichlorohydrine and to be subsequently cured by appropriate cross-linkers. A preliminary screening with a lignin model compound gave the range of the epoxy compounds and the conditions to be tested for the reaction with lignin. Then various experiments with lignin confirmed the potential of 4, 5-epoxytetrahydrophthalic acid diglycidylester, an epoxy compound containing both one cycloaliphatic and two glycidyl type of epoxy groups, as an alternative for epichlorohydrine for the preparation of modified lignin. In the conditions we have defined, the reaction yielded epoxy lignin-based prepolymers as a powder with epoxy level as high as 1,2 mol/kg determined by FTIR spectroscopy. The series of evaluation campaigns in rubber blends showed that the new epoxy-modified lignin associated with p-xylylenediamine can be successfully mixed with natural rubber and enable to reach the properties of our reference mix.

Master of Science / Department of Agricultural Economics / Elizabeth Yeager / The main objective of this analysis was to examine the profitability and efficiency of Kansas farms in order to draw inferences among the profitability, efficiency, and growth of agricultural producers in Kansas. The time period analyzed was 2005 to 2015. Farms in the sample include a mix of 564 crop and/or livestock operations with 11 years of continuous data through the Kansas Farm Management Association data-bank. Efficiency scores were calculated to determine how close each farm was to the production possibilities frontier, or their cost efficiency. Profitability measures, (operating profit margin and return on assets), were obtained for each farm. The profitability dynamics in 2014 and 2015 for these farms changed compared to previous years. Crop farms generated less profits in 2014 and 2015 compared to previous years, and relative profits from average fluctuated more for sampled farms in 2014 and 2015. Farms were also categorized into risk classes. These classifications aim at distinguishing farms that are profitable or not, and their level of solvency, utilizing their net farm income from operations and their debt to asset ratio. Farms are migrating from the low risk classification, showing that Kansas farms are becoming less profitable, but are not transitioning to a higher risk solvency state. These farms will need to focus on utilizing their inputs more efficiently to keep their solvency levels in check. After analyzing persistence in profitability, the results suggest that farms with higher return on assets tend to be more solvent, but farms with higher operating profit margin tend to be less solvent. The analysis also suggests that there might have been persistence in profits in the years prior to 2015. The analysis of relative positioning of farms in terms of return on assets suggests that during 2007-2011 some farms were able to consistently differentiate themselves by generating either below or above normal profits. Some farms were able to become more profitable in 2012 and 2013, while others lagged behind supported by regression results that signaled divergence of profitability levels. The relative positioning analysis for operating profit margin indicates that farms had similar operating profit margins from 2010 through 2014, and divergence occurred in 2015 by farms that were able to differentiate themselves more through the average operating profit margin. Next the efficiencies of the farms were examined. Analysis of the efficiency scores suggests that the cost efficiencies of Kansas farms are not explained by risk classification significantly, but the crop-labor percentage ratio significantly explains the cost efficiency of the farms. The relationship between cost efficiency and profitability measures proved to be the strongest out of the three performance measures due to their correlation. The final step in the analysis was to examine farm characteristics of the top performing farms. Farms were ranked by profitability measurements and the efficiency measure. Variables of interest that were significantly different between the top 25 percent and the bottom 25 percent of farms include total farm assets, value of farm production, crop-labor percentage, crop acres, number of workers, and age of operators.

This thesis studies the concept of Linking Relief, Rehabilitation and Development (LRRD), a topic discussed since the late 1980’s that has failed to be practically implemented, partly because of widely divergent perspectives on the concept. The discourse on LRRD has so far largely been conducted in a top-down fashion with donors constituting the dominant interlocutor, while the perspectives of aid organisations and local communities involved in humanitarian and developmental programmes have been widely overlooked. This thesis thus means to bring clarity to how LRRD is conceptualised by different stakeholders through proposing a comprehensive conceptual framework based from literature, which is used to analyse empirical case studies at the local, national, and international levels. The case studies were conducted in Indonesia and Sweden through interviews with 16 participants and a survey with 20 beneficiaries as respondents. The participants included: beneficiaries at a tsunami post-disaster site, local community leaders, a local level NGO, two national level Indonesia NGOs (MDMC and YEU), and an INGO (Plan International).The research reveal that none of the cases experienced as rigid divide between humanitarian and development action as is often suggested in the literature discourse and through donor policies. All interviewed NGOs expressed that they operated in a way that does create strong humanitarian-developmental linkages and that the major obstacle to achieve this is external pressures, particularly from donor agencies, to operate under exclusively humanitarian or developmental imperatives.

With widespread use of the Internet, data is often shared between organizations in B2B health care networks. Integrating data across all sources in a health care network would be useful to public health surveillance and provide a complete view of how the overall network is performing. Because of the lack of standardization for a common data model across organizations, matching identities between different locations in order to link and aggregate records is difficult. Moreover, privacy legislation controls the use of personal information, and health care data is very sensitive in nature so the protection of data privacy and prevention of personal health information leaks is more important than ever. Throughout the process of integrating data sets from different organizations, consent (explicitly or implicitly) and/or permission to use must be in place, data sets must be de-identified, and identity must be protected. Furthermore, one must ensure that combining data sets from different data sources into a single consolidated data set does not create data that may be potentially re-identified even when only summary data records are created. In this thesis, we propose new privacy preserving data integration protocols for public health surveillance, identify a set of privacy preserving data integration patterns, and propose a supporting framework that combines a methodology and architecture with which to implement these protocols in practice. Our work is validated with two real world case studies that were developed in partnership with two different public health surveillance organizations.

The aim of this project was to analyse the cellular and biomechanical changes after collagen cross-linking (CXL) treatment on postmortem eye-banked human corneas using different UVA intensities and repeated treatments, and to explore the effects of standard collagen cross-linking on keratoconic corneal buttons, in-vitro. Preliminary studies were conducted to assess the feasibility of using eye-banked corneas to assess the effects of collagen cross-linking, and the possibility of applying scanning acoustic microscopy (SAM) to measure the speed of sound/elasticity of corneal tissue. Eye-banked human corneas were successfully cross-linked allowing the effects of CXL to be studied in-vitro and SAM was used effectively to determine the mechanical properties of corneal tissue at different depths. The results of two experiments comparing UVA intensity suggested that no statistically significant difference was found in the histological changes or in the induced stiffness after applying low and high intensity cross-linking on normal human corneas. However, the number of apoptotic cells was found to be significantly less but deeper into the posterior stroma in the high intensity cross-linked corneas. Collectively, these results confirmed the safety and efficacy of both techniques with the advantage of reducing the treatment time using the higher-intensity treatment. In another in-vitro study, keratoconic corneal tissue was used. Different histological and biomechanical outcomes were found between the cross-linked and control keratoconic tissue. The effects of cross-linking were found to penetrate deeper in the keratoconic tissue compared to in the normal corneal tissue found in previous studies. This could be due to the altered collagens and extracellular matrix of the keratoconic corneas, as they were taken from patients in advanced stages of the disease. This study confirmed the importance of having corneal thickness of at least 400μm after epithelial debriding to maintain the endothelial cell density and integrity. Finally, further cross-links were induced when collagen cross-linking treatment was repeated. However, repeating cross-linking three times a deeper cell death close to the endothelium was noticed which suggests that multiple treatments could be unsafe. Additionally, lower speed of sound than the cross-linking twice. This could be due to elimination of the induced cross-links by longer exposure to UVA irradiation. In conclusion, eye-banked human corneas were successfully used to evaluate the effects of cross-linking treatment and repeated treatment. Additionally, keratoconic corneal buttons were used to study the effects of collagen cross-linking in-vitro. This model of using eye-banked human corneas and keratoconic corneal tissue enabled us to study the effects of cross-linking treatment using different protocols and the effects of repeated treatment, and it could ultimately be used to compare the results with in-vivo studies.

Orientadores: Rosiane Lopes da Cunha, Ana Carla Kawazoe Sato / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia de Almentos / Made available in DSpace on 2018-08-18T17:56:49Z (GMT). No. of bitstreams: 1 Costa_AlineAlvaresdaSilva_M.pdf: 2383816 bytes, checksum: 2af9fcfd4c4efd286c4471da25aa7226 (MD5) Previous issue date: 2011 / Resumo: Proteínas são os biopolímeros mais amplamente utilizados nas formulações de emulsões alimentícias como agentes emulsificantes, por serem um aditivo seguro. Com o intuito de entender e melhorar as propriedades emulsificantes das proteínas, inicialmente foram estudadas as macroemulsões óleo-água (O/A) estabilizadas por caseinato de sódio (CN-Na), obtidas em um sistema rotor-estator, sob diferentes concentrações de proteína e condições de pH. Todas as macroemulsões apresentaram separação de fases, devido ao mecanismo de cremeação. Este processo de desestabilização foi reduzido quando existiu o aumento da viscosidade dos sistemas, obtido pela adição de maiores concentrações de proteína e pela redução do pH em direção ao ponto isoelétrico da proteína. Já a utilização da homogeneização a altas pressões promoveu a formação de emulsões cineticamente estáveis, não sendo observada separação de fases com 4% de CN-Na em nenhuma das condições de pH. De modo geral, as emulsões apresentaram comportamento pseudoplástico, com exceção das emulsões estabilizadas com 1% de proteína em pH 5, que se comportaram como fluido dilatante, e em pH 7, como fluido Newtoniano. A redução do pH para os sistemas com 1% de CN-Na levou à desestabilização das emulsões, devido à menor concentração de proteínas adsorvidas não permitirem uma estabilização eletrostática. Foi realizado um tratamento enzimático para melhorar a estabilização das emulsões que separaram de fases em pH ácido. Assim, géis de caseinato de sódio foram reticulados com lacase e ácido ferúlico e as propriedades mecânicas desses géis foram avaliadas. A adição de lacase mediada por ácido ferúlico resultou em géis mais rígidos, firmes e menos deformáveis. As melhores combinações foram selecionadas para o preparo de emulsões O/A estabilizadas com CN-Na com o objetivo de aumentar a estabilidade em pH ácido. O uso desse tratamento enzimático levou a modificações na estrutura da proteína e, com isso, mudanças nas suas propriedades funcionais, o que permitiu o aumento na estabilidade das emulsões. Em geral, essas emulsões tratadas enzimaticamente apresentaram-se mais estáveis, com distribuição de gotas menos polidispersa e com comportamento mais estruturado, apesar do aumento no diâmetro médio das gotas, variando entre 11,79 e 20,17 µm para emulsões em pH 3 contra 6,14 µm medido na emulsão controle (sem tratamento enzimático). Assim, o aumento na estabilidade dessas emulsões deve estar associado ao aumento da viscosidade, que promoveram estabilidade estérica aos sistemas. Portanto, os resultados mostram que foi possível a produção de emulsões ácidas com maior estabilidade a partir do caseinato de sódio, através do tratamento enzimático, originando emulsões com estruturas e propriedades reológicas diferenciadas em comparação com a proteína não reticulada / Abstract: Proteins are biopolymers widely used as a safe additive in the formulation of food emulsions as emulsifying agents. To understand and improve the emulsifying properties of sodium caseinate, initially oil-in-water macroemulsions (O/W) were studied. The emulsions were stabilized by sodium caseinate prepared using a rotor-stator device at different concentrations of protein and pH. All macroemulsions showed phase separation due to the creaming mechanism. This mechanism of destabilization was reduced with the increase of system viscosity, either due to the increase on the concentrations of protein and by the reduction of pH towards to the protein¿s isoelectric point. The use of high-pressure homogenization promoted the formation of stable microemulsions, with no phase separation observed in emulsions with 4% CN-Na. In general, the emulsions exhibited a shear-thinning behavior, except the emulsion containing 1% protein at pH 7, which exhibited Newtonian behavior, and at pH 5, which tended to show a shear-thickening behavior. The reduction of pH in emulsions with 1% CN-Na led to phase separated emulsions, which was attributed to the amount of adsorbed protein, which was insufficient to promote a strong electrostatic repulsion. In order to improve the stabilization of phase separated emulsions at acidic pH, an enzymatic treatment was carried out. Thus, sodium caseinate gels were crosslinked with laccase and ferulic acid and the mechanical properties of these gels were evaluated. The addition of laccase and ferulic acid resulted in gels with increased hardness, firmness and less deformable. The best treatments were selected for the preparation of O/W emulsions stabilized with CN-Na, in order to increase in their stability in acidic pH. The enzymatic treatment caused modifications in the protein structure, resulting in changes of functional properties, which led to an increase in the emulsion stability. In general, these enzymatically treated emulsions were more stable, composed by droplets with lower size distribution and more structured behavior, despite the increased mean droplet diameters (between 11.79 and 20.17 µm for emulsions at pH 3). Thus, the stability increase of these emulsions could be associated to the increase in viscosity, which resulted in steric stability. Results showed that it was possible to produce more stable emulsions containing sodium caseinate in acidic pH using enzymatic treatment, resulting in emulsions with different structures and rheological properties when compared with the non-cross-linked protein / Mestrado / Engenharia de Alimentos / Mestre em Engenharia de Alimentos