Viruses modulate biosynthetic machineries of the host cell for a rapid and efficient virus replication. One important way of modulating protein activity in eukaryotic cells is by reversible phosphorylation. In this thesis we have studied adenovirus and vaccinia virus, two DNA viruses with different replication stategies. Adenovirus replicates and assembles new virions in the nucleus, requiring the host cell transcription and splicing machinieries, whereas vaccinia virus replicates in the cytoplasm, only requiring the cellular translation machinery for its replication. Adenovirus uses alternative RNA splicing to produce its proteins. We have shown that adenovirus takes over the cellular splicing machinery by modulating the activity of the essential cellular SR family of splicing factors. Vaccinia virus, that does not use RNA splicing, was shown to completely inactivate SR proteins as splicing regulatory factors. SR proteins are highly phosphorylated, a modification which is important for their activity as regulators of cellular pre-mRNA splicing. We have found that reversible phosphorylation of SR proteins is one mechanism to regulate alternative RNA splicing. We have demonstrated that adenovirus and vaccinia virus induce SR protein dephosphorylation, which inhibit their activity as splicing repressor and splicing activator proteins. We further showed that the adenovirus E4-ORF4 protein, which binds to the cellular protein phosphatase 2A, induced dephosphorylation of a specific SR protein, ASF/SF2, and that this mechanism was important for regulation of adenovirus alternative RNA splicing. Inhibition of cellular pre-mRNA splicing results in a block in nuclear- to cytoplasmic transport of cellular mRNAs, ensuring free access of viral mRNAs to the translation machinery. We propose that SR protein dephosphorylation may be a general viral mechanism by which mammalian viruses take control over host cell gene expression.

Inicialment, l'interès per aquest treball prové de l'increment de la poblacióimmigrant romanesa i xinesa a Espanya. En tres dècades Espanya ha passat de serun país d'emigració per definició a un país d'immigració, i ara com ara és un delspaïsos de la Unió Europea que més estrangers ha rebut en els últims anys. Enaquest context, la present Tesi Doctoral és un estudi sobre adquisició de segonesllengües, en el qual s'analitzen els usos correctes i les interferències com a fontd'errors en l'ús de l'L2 en comparació amb l'L1 i, en conseqüència, pretén ser unaaportació al coneixement acumulat en l'àmbit de la lingüística aplicada. Hemseleccionat un total de 60 alumnes de 12, 14 i 16 anys (12 autòctons i 48 alumnesimmigrants) escolaritzats en 6è de Primaria i en 2n i 4t d'Educació SecundàriaObligatòria, dividits en dos grups, en funció del temps d'estada al país d'acollida.Per tal de complir amb aquest propòsit, hem escollit com a objecte d'estudi latransferència lingüística de l'L1 a l'L2 en el camp de la morfosintaxi. D'aquestamanera, l'objectiu general del estudi és analitzar la manera en la qual es dóna lainfluència de l'L1 en l'adquisició de l'L2 en el cas de l'alumnat romanès i xinèsescolaritzat a Catalunya a partir dels usos correctes i incorrectes en la flexiónominal i verbal, emfasitzant les possibles diferències entre aquests parlants ambllengües maternes molt distants i interpretar-los per tal d'aportar elements per a lamillora de les pràctiques educatives i els processos d'ensenyament-aprenentatge dellengües. / Desde el principio, el interés por este trabajo ha surgido debido al incremento de lapoblación inmigrada rumana y china en España. En tres décadas, España ha pasado deser un país de emigración por definición a un país de inmigración, y hoy por hoy es unode los países de la Unión Europea que más extranjeros ha recibido en los últimos años.En este contexto, la presente Tesis Doctoral es un estudio sobre adquisición de segundaslenguas, en el que se analizan los usos correctos y las interferencias como fuente deerrores del uso de la L1 comparada con la L2, y, en consecuencia, pretende ser unaaportación al conocimiento acumulado en el ámbito de la lingüística aplicada. Hemosseleccionado un total de 60 alumnos de 12, 14 y 16 años (12 autóctonos y 48 alumnosinmigrantes) escolarizados en 6º de Primaria y en 2º y 4º de Educación SecundariaObligatoria divididos en dos grupos en función del tiempo de estancia en el país deacogida. Para cumplir con este fin, hemos elegido como objeto de estudio latransferencia lingüística de la L1 a la L2 en el campo de la morfosintaxis. De estamanera, el objetivo general de nuestro estudio es analizar la manera en que se da lainfluencia de la L1 en la adquisición de la L2 en el caso del alumnado rumano y chinoescolarizado en Cataluña a partir de los usos correctos e incorrectos en la flexiónnominal y verbal, enfatizando las posibles diferencias entre estos hablantes de lenguasmaternas muy distantes e interpretarlos con tal de aportar elementos para una mejora delas prácticas educativas y los procesos de enseñanza-aprendizaje de las lenguas. / The interest in this paper came along with the increasing number of Romanian andChinese immigrants in Spain. Over the last three decades, Spain has converted from acountry of emigrants into a country of immigrants, being nowadays one of the countriesin European Union with a significant influx of foreigners. In this context, the herebyPh.D. thesis is a research in Second Language Acquisition, grounded on the study ofcorrect uses and linguistic interferences as source of errors in the use of native language(L1) compared to the second language (L2), and thus, it attempts to bring a significantcontribution in the field of applied linguistics. Our study was conducted on 60 studentsof 12, 14 and 16 years old (12 natives and 48 immigrant students) from the 6th grade(Primary Education), and from the 2nd and 4th grade of Secondary Education divided ingroups, depending on the length of their stay in Spain. In order to successfully reach thisobjective, I selected as research theme the morphological and syntax transfer from L1 toL2. Thus, the overall objective of our study is to analyze how L1 influences theacquisition of L2 in the case of Romanian and Chinese students attending school inCatalunya focusing on the correct and incorrect use of language in the nominal andverbal inflexions, by emphasizing possible differences between these speakers ofdifferent L1 languages, in order to assess them and to improve the educational practiceand teaching / learning of foreign languages.

L'étude de nombreux phénomènes astronomiques repose sur la recherche de périodicités dans des séries temporelles (courbes de lumière ou de vitesse radiale). En raison des contraintes observationnelles, la couverture temporelle des données résultantes est souvent incomplète, présentant des trous périodiques ainsi qu'un échantillonnage irrégulier. L'analyse du contenu fréquentiel de telles séries basée sur le spectre de Fourier s'avère alors inefficace et les méthodes heuristiques de déconvolution de type CLEAN, couramment utilisées en astronomie, ne donnent pas entière satisfaction. Cette thèse s'inscrit dans le formalisme fréquemment rencontré depuis les années 1990 abordant l'analyse spectrale sous la forme d'un problème inverse, le spectre étant discrétisé sur une grille fréquentielle arbitrairement fine. Sa régularisation est alors envisagée en traduisant la nature a priori parcimonieuse de l'objet à reconstruire: nous nous intéressons ici à la recherche de raies spectrales. <br />Une première approche envisagée a trait au domaine de l'optimisation et consiste à minimiser un critère de type moindres carrés, pénalisé par une fonction favorisant les solutions parcimonieuses. La pénalisation par la norme l1 est en particulier étudiée en extension à des variables complexes et s'avère satisfaisante en termes de modélisation. Nous proposons des solutions algorithmiques particulièrement performantes permettant d'envisager une analyse à très haute résolution fréquentielle. <br />Nous étudions ensuite la modélisation probabiliste des amplitudes spectrales sous la forme d'un processus Bernoulli-Gaussien, dont les paramètres sont estimés au sens de la moyenne a posteriori à partir de techniques d'échantillonnage stochastique, permettant d'envisager une estimation totalement non supervisée. L'interprétation probabiliste du résultat ainsi que l'obtention conjointe des variances associées, sont alors d'un intérêt astrophysique majeur, s'interprétant en termes de niveaux de confiance sur les composantes spectrales détectées. Nous proposons dans un premier temps des améliorations de l'algorithme échantillonneur de Gibbs permettant d'accélérer l'exploration de la loi échantillonnée. Ensuite, nous introduisons des variables de décalage fréquentiel à valeur continue, permettant d'augmenter la précision de l'estimation sans trop pénaliser le coût calculatoire associé. <br />Pour chaque méthode proposée, nous illustrons sur des simulations la qualité de l'estimation ainsi que les performances des algorithmes développés. Leur application à un jeu de données issu d'observations astrophysiques est enfin présentée, mettant en évidence l'apport d'une telle méthodologie par rapport aux méthodes d'analyse spectrale habituellement utilisées.

In this research, Communicative Accommodation Theory (CAT) is investigated while native speakers address nonnative peers. For the intentions of this research, three native speakers of Canadian English were asked to have conversations with native and nonnative peers. The conversations were in the form of giving directions on the map. Later on, the participants’ formants and vowel durations were measured and used for comparing native-nonnative peer effect(s) on the speakers’ vowel formants and duration. Based on the analyses, it is suggested that accommodation may take place based on providing stereotypical vowel durations and formants, as well as reducing inter-token variations in the nonnative peer context.

The paper is concerned with the impact of market research prior to integration, on the structures of noncompetitive industries in integrated economy. The analysis focuses on separated, single commodity, monopolistic markets with stochastic demand. Monopolistic firms are considered in dynamic multiperiod model, where intertemporal links are determined by expenditures on market research in a present period and benefits from this activity (i.e., smaller variance of the prediction error) in the future. Assuming that each firm maximizes its total discounted expected utility from profit in indefinite time, we show that the optimal market research strategy is stationary and depends on market size. Consequently, in the period following integration firms operating prior to integration in small markets (such as Slovenia, Czech Republic, Hungary or Estonia) are expected to have much less information about the integrated market than their competitors operating before integration on European market. This informational asymmetry may affect the structure of the industry in integrated economy. In the extreme case, the firm operating before integration in the small market can be ruled out from the integrated market. (authors' abstract)

In a series of work initiated by Nemirovsky and Yudin, and later extended by Nesterov, first-order algorithms for unconstrained minimization with optimal theoretical complexity bound have been proposed. On the other hand, conjugate gradient algorithms as one of the widely used first-order techniques suffer from the lack of a finite complexity bound. In fact their performance can possibly be quite poor. This dissertation is partially on tightening the gap between these two classes of algorithms, namely the traditional conjugate gradient methods and optimal first-order techniques. We derive conditions under which conjugate gradient methods attain the same complexity bound as in Nemirovsky-Yudin's and Nesterov's methods. Moreover, we propose a conjugate gradient-type algorithm named CGSO, for Conjugate Gradient with Subspace Optimization, achieving the optimal complexity bound with the payoff of a little extra computational cost. We extend the theory of CGSO to convex problems with linear constraints. In particular we focus on solving $l_1$-regularized least square problem, often referred to as Basis Pursuit Denoising (BPDN) problem in the optimization community. BPDN arises in many practical fields including sparse signal recovery, machine learning, and statistics. Solving BPDN is fairly challenging because the size of the involved signals can be quite large; therefore first order methods are of particular interest for these problems. We propose a quasi-Newton proximal method for solving BPDN. Our numerical results suggest that our technique is computationally effective, and can compete favourably with the other state-of-the-art solvers.

Un des mécanismes d'évasion des cellules tumorales au système immunitaire fait intervenir la famille des molécules de type B7. Ces molécules de costimulation ont aussi un rôle dans les mécanismes de tolérance immunitaire. La molécule B7-H1 (PD-L1 ou CD274) inhibe directement les lymphocytes T cytotoxiques (CTL) et est fortement exprimée à la surface de nombreux types de cellules tumorales. Les mécanismes de régulation de son expression ne sont pas bien connus. Il existe une augmentation d'expression de cette molécule après stimulation par l'IFNγ et récemment les ligands des Toll-Like-Receptor (TLR) 2, 4 et 9 ont été impliqués dans sa régulation au niveau du myélome multiple. L'implication possible des TLR dans l'expression de B7-H1 suggère un rôle possible de pathogènes dans l'échappement des tumeurs au système immunitaireIl n'existe aucune donnée dans la littérature sur l'expression des TLR au niveau des cellules blastiques de patients atteints de leucémies aigues myéloblastique (LAM). Nous avons dans un premier temps étudié l'expression des TLR dans les LAM et l'inductibilité de l'expression de la molécule B7-H1 par les ligands des TLR en cytométrie en flux. Nos résultats montrent que les TLR sont exprimés dans les LAM, avec une grande variabilité suivant les patients. On observe une augmentation significative de l'expression de B7-H1 après stimulation par le LPS (ligand TLR4) alors qu'elle n'est pas significative pour les autres ligands (PGN et ODN). Comme dans les tumeurs solides et le myélome multiple, l'expression de B7-H1 est augmentée par l'IFN γ. Des tests de lyse CTL sont en cours afin de confirmer le rôle fonctionnel de cette expression de B7-H1 via les TLR dans les LAM.Une autre partie de l'étude a été réalisée sur deux lignées murines leucémiques : DA1-3B et WEHI-3B, afin de disposer de modèles expérimentaux d'expression de B7-H1. On retrouve sur ces deux modèles l'augmentation d'expression de B7-H1 après stimulation par l'IFN γ et les ligands des TLR. L'utilisation de différents inhibiteurs chimiques des voies de signalisation suggère le rôle des voies MEK/ERK et de la voie JAK/STAT dans l'expression de cette molécule. La voie PI3kinase/Akt semble au contraire jouer un rôle inhibiteur. Le travail se poursuit avec la transfection de transdominants négatifs des différentes voies et de mutants constitutivement actifs. L'objectif à terme est de tester des stratégies d'immunothérapies des LAM par blocage pharmacologique de l'expression de B7-H1.

In this research, Communicative Accommodation Theory (CAT) is investigated while native speakers address nonnative peers. For the intentions of this research, three native speakers of Canadian English were asked to have conversations with native and nonnative peers. The conversations were in the form of giving directions on the map. Later on, the participants’ formants and vowel durations were measured and used for comparing native-nonnative peer effect(s) on the speakers’ vowel formants and duration. Based on the analyses, it is suggested that accommodation may take place based on providing stereotypical vowel durations and formants, as well as reducing inter-token variations in the nonnative peer context.

A surface reconstruction and image enhancement non-linear finite element technique based on minimization of L1 norm of the total variation of the gradient is introduced. Since minimization in the L1 norm is computationally expensive, we seek to improve the performance of this algorithm in two fronts: first, local L1- minimization, which allows parallel implementation; second, application of the Augmented Lagrangian method to solve the minimization problem. We show that local solution of the minimization problem is feasible. Furthermore, the Augmented Lagrangian method can successfully be used to solve the L1 minimization problem. This result is expected to be useful for improving algorithms computing digital elevation maps for natural and urban terrain, fitting surfaces to point-cloud data, and image super-resolution.

Fluent sentence comprehension requires the automatic application of grammatical principles, in combination with other kinds of information, to a linear input string. The latter can vary greatly in complexity, and it has been observed that structures that involve non-adjacent relationships (non-local dependencies) cause particular processing difficulties, for even fluent readers. This thesis focuses on the potential role that L1 transfer plays in processing non-local dependencies in the L2. Although L1 transfer has been demonstrated in a variety of L2 domains (Juffs, 2005), the degree to which L1 syntactic properties influence the L2 in real-time processing remains an open question (Clahsen & Felser, 2006b). A better understanding of L1 influence on L2 processing has important implications for emerging models of L2 sentence processing and SLA theory in general. This thesis examines L1 effects on the processing of L2 English relative clauses. It builds on established models of L1 relative clause processing (e.g., Gibson, 1998) and focuses on the asymmetry observed in the processing of subject versus object relative clauses. Research has shown that L1 individuals find object relative clauses more difficult to process than their subject counterparts (King & Just, 1991), and off-line L2 research suggests that this asymmetry might hold for L2 processing as well (Izumi, 2003; Kanno, 2007). This thesis will examine the asymmetry on-line and investigate the role that L1 background plays in on-line performance, both alone and in combination with other processing factors. The main data are on-line performance by three groups of advanced-level learners of L2 English from Japan, China, and Vietnam. The three L1s differ from English to varying degrees in relative clause formation, as reflected in head direction, branching direction and the presence of an overt relative marker. Vietnamese shares all three properties with English and Japanese none, leading to the prediction that the Japanese learners will have the greatest difficulty, as evident in the largest subject/object asymmetry in processing times. Evidence for the effect of L1 syntactic properties on L2 processing will be sought in a set of studies that compare processing outcomes for these learners across key individual difference factors in L2 processing, working memory capacity, lexical processing efficiency, and proficiency. The effect of non-structural information, in the form of animacy contrasts and frequency effects, on L2 syntactic processing across the L1 groups will also be examined. By systematically examining group differences across these key factors, the degree and nature of potential L1 influence on the processing of these non-local dependencies can be better assessed. Three groups of advanced-level L2 English learners from China, Japan, and Vietnam, a group of intermediate-level learners in Japan, and native English controls participated in the data collection. Participants were first tested on a battery of individual difference measures that established language proficiency and general processing skill for individual participants and groups. On-line reading data of relative clause structures were then collected and reported in a set of five studies. Study 1 examined reading when the target structures contained high frequency content words, which minimized word recognition demands. Advanced L2 learners took longer to read object relative clauses, and the difference was the greatest for the Japanese group. Study 2 revealed that the subject/object asymmetry and L1 influence evident in Study 1 disappeared when word recognition was effortful, as in sentences with low frequency words, suggesting that L1 syntactic influence may only be evident when lexical processing is relatively automatic. Study 3 examined the effect of proficiency differences on processing outcomes by comparing performance by Intermediate and Advanced Japanese learners. The Intermediate learners showed no subject/object asymmetry in sentences with high frequency words, providing further evidence that processing efficiency is a key factor when considering possible L1 effects in L2 processing. Study 4 demonstrated that advanced L2 learners benefited from animacy cues in on-line processing, with the difficulty associated with object relative sentences and the L1 effect evident in sentences with animate head nouns neutralized in sentences with inanimate head nouns. The findings indicate that L2 learners may rely on lexical information as an alternative to syntactic knowledge in some cases. Finally, Study 5 compared processing performance on full or reduced relative clauses to examine both the influence of L1 syntactic properties and L1 structural frequency biases on L2 processing outcomes. Although the influence of frequency information was not evident, the results suggested that the availability of an overt relative marker in the L1 affected L2 processing outcomes. Overall, the results provide some support for the view that L1 background influences the processing of complex syntactic structures in the L2, a finding consistent with Juffs (2005) but contrary to Clahsen et al. (Clahsen & Felser, 2006b). However, the effect is sensitive to a range of factors, which appears to support the claim by the latter that L2 learners may rely more on lexical-semantic information than L1 individuals.