No description available.

No description available.

A recently developed group-sequential response-adaptive design to compare two treatments with immediate normally distributed responses and known variances is considered. The power function of the test is the same as that under non-adaptive sampling, and significant decreases in the inferior treatment number can be achieved with only minor increases in the average sample number. Reasonably accurate corrected confidence intervals for both the treatment mean difference and the individual means are obtained by constructing approximately pivotal quantities. An approximation to the bias of the maximum likelihood estimator of the treatment mean difference is also studied. When the variances of the response variables are unknown, inaccurate estimates of these can affect the Type II error rate considerably. A new modified version of an existing sample size re-estimation method is developed for group-sequential response-adaptive designs for normal data with unknown variances. The principal modifications involve updating the required sample size at each interim analysis and calculating the test statistic based on current estimates of the variances. Simulation is used to compare the performance of this test with modified versions of two other tests from the recent literature. The power is shown to be more accurately maintained in the new test. An analogous group-sequential response-adaptive design to compare two treatments with immediate dichotomous responses is then developed. Since the variances of the response variables are unknown in binary response trials, due to their dependence on the unknown success probabilities, the new sample size re-estimation method is incorporated into the design. Two parameters of interest are considered, the log odds ratio and the simple difference between the probabilities of success. Three adaptive urn models are studied and their properties are compared to a sequential maximum likelihood estimation rule that minimises the expected number of treatment failures. Simulation results favour the drop-the-loser rule

No description available.

Proefschrift Universiteit van Amsterdam. / Met lit. opg. en een samenv. i.h. Nederlands.

No description available.

Today, some kind of laboratory results is required for around 70% of the diagnostics and follow-ups for diseases. In many of the cases the time from sampling to a result is very critical. Therefore the discussion of how to improve this situation has begun. For many analyses serum has been the routine choice for a long time but now it is disputed. After blood collection in a serum tube it is essential to wait 30-60 minutes before centrifugation and analysis of the sample, a long time for someone in an acute state. Other problems like post centrifugation clots of fibrin causing false results or time-consuming reruns of the sample have also been reported. These problems have initiated the laboratory in Hudiksvall’s hospital to find out an alternative to the common serum sampling.In this report, the differences between serum and lithium heparin plasma for 31 analyses has been evaluated. Paired blood samples, one serum and one plasma, were collected for routine, hormonal and for tumor markers analyses and analyzed in a Cobas c501, e411 or e601 (ROCHE). The results of the analyzed samples were compared to each other by statistical analysis.The results prove that serum and lithium heparin plasma is equal for ALT, GGT, NT-proBNP, FT3, FT4, cobalamin, LH, prolactin, TSH, CA19-9, CEA and PSA. The results also prove that serum and lithium heparin plasma is not equal for 19 other analyses. Therefore, a shift between different types of sampling is not to be recommended without further evaluations.

Il existe un certain nombre de situations au cours desquelles l’intégrité du génome cellulaire est mise en danger. Ceci se produit notamment lors de mouvements de gènes (translocation, éléments mobiles), lors d’infections virales parfois intégratives (AAV, HBV, HPV) ou lors d’infections rétrovirales. En effet, le cycle de réplication des rétrovirus nécessite une étape d’intégration du génome viral dans l’ADN génomique de la cellule infectée. Malgré les nombreuses études menées sur les infections par le VIH, les cancers viro-induits ou non et les thérapies géniques basées sur les rétrovirus, aucune donnée n’est actuellement disponible sur les modifications cellulaires induites par de telles perturbations chromosomiques. L’objectif du projet était d’identifier des mécanismes cellulaires induits par des atteintes à l’intégrité du génome, plus précisément par l’insertion de molécules d’ADN non cellulaire dans les chromosomes. L’intégration de matériel génétique additionnel a été induite par des vecteurs lentiviraux dérivés du VIH-1. Les modifications cellulaires uniquement dues à l’intégration ont été isolées par comparaison de vecteurs intégratif et non intégratif. Des cellules primaires du derme humain ont été sélectionnées pour l’étude. Les temps post infection et les doses virales les plus adéquates ont été sélectionnés grâce à des expériences de cinétiques d’intégration couplées à une quantification des ADN viraux intégrés. L’analyse des modifications cellulaires induites par l’intégration de l’ADN étranger a portée sur l’ensemble du transcriptome et sur l’ensemble du protéome cellulaire. Dans le but d’effectuer une analyse transcriptomique, des puces à ADN, représentant le génome humain complet, ont été utilisées. Cette étude a démontré une forte répression transcriptionnelle induite par l’intégration. De plus, toutes les fonctions cellulaires sont perturbées par le processus. Finalement, une classification par interactions et fonctions biologiques a mis en évidence cinq processus cellulaires majoritairement affectés par l’intégration de l’ADN étranger, qui correspondent au cycle et à la mort cellulaire, au remodelage et à la réparation de la chromatine et à la réponse immunitaire ou au stress. Dans le but de compléter cette analyse transcriptomique, une étude protéomique a été réalisée. Les protéines cellulaires ont été séparées sur des gels bi-dimensionnels. Parmi les neuf protéines identifiées en spectrométrie de masse, certaines appartiennent au cytosquelette et d’autres aux mécanismes de réponse au stress. L’intégration d’un ADN étranger au sein du génome provoque donc bien des perturbations cellulaires. L’intégration de matériel génétique additionnel ne concernant pas seulement les rétrovirus, les données obtenues lors de cette étude pourront permettre (i) de développer des stratégies de défenses contre les rétrovirus ou contre les autres maladies caractérisées par des atteintes à l’intégrité du génome, (ii) d’évaluer les risques encourus par l’intégration d’un vecteur thérapeutique dans le génome cellulaire lors des thérapies géniques ou des expériences de transfert de gènes / In numerous situations, cell genome integrity could be in danger. This is the case during gene movements (translocations, mobiles elements), during viral infections that could be sometimes integrative (AAV, HBV, HPV) or during retroviral infections. Indeed, the replication cycle of retroviruses requires a viral genome integration step. In spite of the studies on HIV infections, viral or non viral cancers and retrovirus-based gene therapy, no data are available concerning the cellular modifications induced by such chromosomal disruptions. The aim of work was to identify cellular mechanisms induced by the insertion of a non cellular DNA into the chromosomes. The integration of additional DNA was provoked by HIV-1-based lentiviral vectors. Cellular modifications only due to the integration step were isolated by comparison of integrative and non integrative vectors. Primary human dermal fibroblasts cells were selected for the study. Optimal times post infection and viral quantities were defined using kinetic integration experiments and integrated viral DNA quantifications. The study of cellular modifications induced by the integration of the foreign DNA was applied on the cellular global transcriptome and proteome. In order to perform a transcriptomic analyses, DNA microarray corresponding to the whole human genome, were used. This study revealed a strong transcriptional repression induced by the integration. Moreover, every cellular function are disturbed by the process. Finally, a network based on molecular interactions and biological functions underlined five cellular processes mostly affected by the foreign DNA integration and corresponding to the cell cycle and death, the remodelling and repair of chromatin and the immunity or stress responses. To complete this transcriptomic analyses, a proteomic study was realized. Cellular proteins were separated on 2D gels. Among the nine proteins identified by mass spectrometry, some are linked to the cytoskeleton and other to the cellular stress. Thus, integration of a foreign DNA into the genome provoked cellular perturbations. As additional DNA integration do not only concern retroviruses, data obtained during this study could allow (i) the development of defensive strategies against retroviruses or other diseases implicating the genome integrity and (ii) the evaluation of risks linked to the integration of a therapeutic vector into the genome during gene therapy and gene transfer experiments

Cette thèse s'inscrit dans le cadre de la bioinformatique. Les outils mathématiques les plus utilisés dans ce travail relèvent de la théorie des graphes, des statistiques, de la théorie des ensembles et des mathématiques discrètes. Ces mathématiques ont permis de développer des modèles de systèmes biologiques ainsi que des algorithmes efficaces dans l'étude concrète de ces modèles. La nécessité d'analyses de jeux de données de très grande taille a rendu critique dans notre démarche cette notion d'efficacité des algorithmes. Il faut enfin remarquer que le champ biologique qui a servi de support à cette thèse nous a conduit à explorer un domaine particulier au sein de la théorie de la complexité, à savoir le développement et l'analyse des algorithmes d'énumération [etc...] / In the results presented in the present manuscripts, graph theory and combinatorial optimizationtecniques, have been used to model and solve biological problems. The manuscript is divided in twoparts, each one containing the mathematical and biological background of a given application and ouroriginal contributions to it.Part I groups a set of results designed for phylogenetics analysis, and in particular for reconstructingthe co-evolution of two groups of organisms (the so called co-phylogeny reconstruction problem).Although the addressed problem was treated in the available there was no method that solved suchproblem in a complete and efficient way. We thus developed and implemented a new one, calledEucalypt, with this purpose in mind. This not only provides a novel and usable software for cophylogenyreconstruction but also allows to investigate how the event-based model performs inpractice in terms of thenumber and quality of the solutions obtained. We compared our method to the available software. Bylooking at the results obtained, some interesting considerations about the advantages anddisadvantages of the commonly accepted mathematical model could be drawn. Finally, we introduceda new version of the problem where the host-switches are distance bounded: the k-bounded-All-MPRproblem. Eucalypt solves both problems in polynomial delay. These results have been accepted forpublication by the jounal Algorithms for Molecular Biology. The relative software is publicyavailable.Our studies show that the 'most parsimonious scenario' approach presents some limitationsthat cannot be ignored. To deal with these problems, we developed a second algorithm, called Coala,based on an approximate Bayesian computation approach for estimating the frequency of the events.The benefits of this method are twofold: it provides more confidence in the set of costs to be used in areconciliation, and it allows to estimate the frequency of the events in the cases where a reconciliationmethod cannot be applied. These results are currently under review by the jounal Systematic Biology.The relative software is publicy available.In Part 2 another set of studies is presented. Our original model for the contig scaffolding problem,and our algorithm MeDuSa, are presented and tested. Unlike traditional software, it does not rely eitheron paired-end information of sequencing reads or on a phylogenetic distance of the microorganismsused in the analysis. This drastically increases the usability of our software and, at the same time,reduces the computational time required for genome scaffolding. We show that the algorithmimplemented in MeDuSa, in most cases, is capable of producing less and longer scaffolds incomparison to commonly used scaffolders, while maintaining high accuracy and correctness of thepredicted joins. These results are currently under revision by the journal Bioinformatics.Finally, during the development of this method we encountered some pure theoretical open problemsand we decided to dedicate part of our job to their analysis. The last chapter is then dedicated to a setof problems, all related to the Implicit Hitting set enumeration problem. After some formal definitions,an original NP-completeness result is presented and the future directions of our work are described

<p>Two species of Chrysaora are described from the northern Benguela ecosystem: C. fulgida and C. africana. These species can be diagnosed by a combination of morphological features including lappet and tentacle number, shape of lappets, colouration patterns (alive), shape of the proximal portion of radial septa, gastrovascular pouch shape, point of attachment of gonads and the presence or absence of small raised nematocyst warts on the exumbrellar surface. Objective, quantitative statistical analyses coupled with molecular sequence data support the qualitative morphological dissimilarity observed, as these analyses unambiguously diagnose C. fulgida and C. africana as two distinct species. There is a strong superficial resemblance between the C. fulgida material described here and the preserved specimens of C. hysoscella examined at the Natural History Museum, London. Thorough investigation does however allow the separation of these two species. Morphological features found to be dissimilar were the proximal portion of the manubrium, gastrovascular pouch shape and the presence or absence of sperm sacs. Objective, quantitative statistical analyses support these findings. Nuclear sequence variation suggests considerable divergence between the two species but additional molecular work is needed.</p>