Les courbes de tarage complexes, qui prennent en entrée la hauteur d'eau et des variables supplémentaires, sont nécessaires pour établir les chroniques de débit des cours d'eau là où la relation hauteur-débit n'est pas univoque. Dans le même cadre bayésien, des méthodes à base hydraulique sont proposées et testées pour construire les courbes de tarage complexes et estimer leurs incertitudes : des modèles hauteur-gradient-débit (SGD) pour résoudre l'hystérésis due aux écoulements transitoires, des modèles hauteur-dénivelée-pente (SFD) pour résoudre le remous variable aux stations à double échelle, le modèle hauteur-période-débit (SPD) pour résoudre les détarages nets dus aux évolutions du lit. Chaque modèle a été appliqué à des stations hydrométriques variées et évalué grâce à des analyses de sensibilité. Pour chacune des trois sources de non-univocité de la relation hauteur-débit, les méthodes bayésiennes proposées fournissent non seulement une analyse d'incertitude quantitative mais aussi des solutions efficaces à des problèmes récurrents que posent les procédures traditionnelles pour les courbes de tarage complexes. / Complex rating curves, with stage and additional variables as inputs are necessary to establish streamflow records at sites where the stage-discharge relation is non-unique. Within the same Bayesian framework, hydraulically-based methods are introduced and tested to develop complex rating curves and estimate their uncertainties: stage-gradient-discharge (SGD) models to address hysteresis due to transient flow, stage-fall-discharge (SFD) models to address variable backwater at twin gauge stations, stage-period-discharge (SPD) model to address net rating changes due to bed evolution. Each model was applied to contrasting hydrometric stations and evaluated through sensitivity analyses. For each of the three sources of non-uniqueness in the stage-discharge relation, the proposed Bayesian methods provide not only quantitative uncertainty analysis but also efficient solutions to recurrent problems with the traditional procedures for complex ratings.

INTRODUÇÃO: O subgrupo Anopheles strodei é pouco estudado apesar de sua importância epidemiológica potencial. Espécies desse subgrupo foram encontradas naturalmente infectadas por parasitos que causam malária em humanos, Plasmodium falciparum, Plasmodium vivax e Plasmodium malariae, no Brasil. O subgrupo Anopheles strodei compreende oito espécies: An. rondoni (Neiva & Pinto), An. albertoi Unti, An. arthuri Unti, An. strodei Root, An. strodei CP, e três outras espécies que foram propostas por Bourke et al. (2013), mas não foram descritas: An. arthuri B, An. arthuri C e An. arthuri D. OBJETIVOS: A definição e delimitação precisa de espécies que atuam como vetores de agentes infecciosos é um dos objetivos da entomologia da saúde pública. Os objetivos deste estudo foram: 1) Estabelecer as relações filogenéticas entre espécies do Subgrupo Strodei; 2) Estimar a distribuição espacial potencial das espécies do Subgrupo Strodei; 3) Confirmar a presença de quatro espécies sob o nome An. arthuri. MÉTODOS: Sequências de DNA de um gene mitocondrial (fragmento de 658 pares de bases do código de barras do gene COI, citocromo oxidase subunidade I) e de três genes nucleares codificadores de proteínas (White, CAD e CAT) foram empregadas para estabelecer as relações filogenéticas potenciais entre as espécies que compõe o subgrupo An. strodei. As análises filogenéticas foram conduzidas utilizando abordagem Bayesiana das sequências de DNA dos quatro genes. Para estabelecer a distribuição espacial potencial das espécies, utilizou-se abordagem de máxima entropia de nichos ecológicos. Para isso as localidades das coletas, juntamente com os dados climáticos e geográficos foram introduzidos no programa MAXENT. RESULTADOS: Os resultados das análises filogenéticas demonstraram e, portanto, confirmaram o monofiletismo do Subgrupo Strodei, a presença de pelo menos sete espécies sob o nome An. strodei, ou seja, corroborou a validade de An. albertoi, An. arthuri, An. strodei, An. strodei CP, além das espécies denominadas, preliminarmente, como An. arthuri B, An. arthuri C e An. arthuri D. Portanto, como definida atualmente, An. arthuri não representa grupo monofilético, pois inclui táxons que deverão ser formalmente descritos em estudos futuros. CONCLUSÃO: As distribuições potenciais de espécies do Subgrupo Strodei foram propostas pela primeira vez. Cinquenta e cinco sequências do gene nuclear CAT e outras 46 sequências do gene nuclear CAD únicas foram recentemente caracterizadas para espécies do Subgrupo Strodei de Anopheles (Nyssorhynchus), confirmando a presença de pelos menos sete espécies, além de An. rondoni que não foi alvo deste estudo, mas de outros anteriores que confirmaram a validade da mesma. / Introduction Anopheles strodei sensu lato is an understudied subgroup of potential epidemiological importance, having been found naturally infected in Brazil with Plasmodium falciparum, Plasmodium vivax and Plasmodium malariae. An. strodei s.l. is currently composed on 8 species: An. albertoi Unti, An. CP Form, An. rondoni (Neiva & Pinto), An. strodei Root, An. arthuri Unti and three other unnamed species that have been proposed by Bourke et al. (2013): An. arthuri B, An. arthuri C and An. arthuri D. Objectives As delineating species accurately is an essential goal of public health entomology, the objectives of this study were to: 1) Determine the phylogenetic relationships within the Strodei Subgroup and reaffirm or reject the hypothesis of the 3 new species (An. arthuri B, An. arthuri C and An. arthuri D) 2) Address the potential spatial distribution of species of the An. strodei subgroup to provide support for the candidate species in the Strodei Subgroup Methods Bayesian inference, which included DNA sequences of one mitochondrial and three nuclear protein coding genes: CO1, white, CAD and CAT, was used to determine the phylogenetic relationship within the group. To propose a species distribution, collection localities, along with climatic and geographic data were input into MAXENT. Results When analyzing the four molecular markers employed, support was found for allopatry in the Strodei Subgroup. The paraphyletic clade of An. arthuri was supported. Conclusion Potential species distributions of the Strodei Subgroup were addressed for the first time. Fifty-five unique CAT sequences and 46 unique CAD sequences were newly characterized.

I consider statistical modelling of data gathered by photographic identification in mark-recapture studies and propose a new method that incorporates the inherent uncertainty of photographic identification in the estimation of abundance, survival and recruitment. A hierarchical model is proposed which accepts scores assigned to pairs of photographs by pattern recognition algorithms as data and allows for uncertainty in matching photographs based on these scores. The new models incorporate latent capture histories that are treated as unknown random variables informed by the data, contrasting past models having the capture histories being fixed. The methods properly account for uncertainty in the matching process and avoid the need for researchers to confirm matches visually, which may be a time consuming and error prone process. Through simulation and application to data obtained from a photographic identification study of whale sharks I show that the proposed method produces estimates that are similar to when the true matching nature of the photographic pairs is known. I then extend the method to incorporate auxiliary information to predetermine matches and non-matches between pairs of photographs in order to reduce computation time when fitting the model. Additionally, methods previously applied to record linkage problems in survey statistics are borrowed to predetermine matches and non-matches based on scores that are deemed extreme. I fit the new models in the Bayesian paradigm via Markov Chain Monte Carlo and custom code that is available by request.

This quantitative study assessed the association of the design methods used for early phase oncology studies (adaptive versus traditional) and the outcome of late stage clinical trials. Differences by cancer type and by drug classification were also assessed. The theoretical and conceptual frameworks used were the general systems theory and the design and evaluation of complex interventions, respectively. Units of analysis were individual oncology studies in the ClinicalTrials.gov database and Bayesian logistic modeling was applied on a random sample of 381 studies initiated after November 1999 to December 2016. When assessing study design and outcome, there were lower odds of a positive outcome when adaptive methods were used though this association was not statistically significant (OR [95% highest posterior density (HPD)]:0.66 [0.20, 1.21]). Among the different drug types, using adaptive compared to traditional methods was associated with significantly higher odds of a positive outcome for taxanes, OR: 2.75, 95% HPD: 1.01, 5.16) and other, OR: 3.23, 95% HPD: 1.58, 5.46) but no association among studies of monoclonal antibodies or protein kinase inhibitors. Also, there were no significant associations between early phase study design and outcome in late phase studies by cancer type (lung, breast, other). Further research should be conducted using all completed oncology clinical trials in the database to more precisely determine the relationship between adaptive study design in early phase oncology studies and outcomes in late stage studies. Social change can occur through increased uptake of adaptive design methods, which may lead to more efficacious cancer treatment options.

In the present paper we study switching state space models from a Bayesian point of view. For estimation, the model is reformulated as a hierarchical model. We discuss various MCMC methods for Bayesian estimation, among them unconstrained Gibbs sampling, constrained sampling and permutation sampling. We address in detail the problem of unidentifiability, and discuss potential information available from an unidentified model. Furthermore the paper discusses issues in model selection such as selecting the number of states or testing for the presence of Markov switching heterogeneity. The model likelihoods of all possible hypotheses are estimated by using the method of bridge sampling. We conclude the paper with applications to simulated data as well as to modelling the U.S./U.K. real exchange rate. (author's abstract) / Series: Forschungsberichte / Institut für Statistik

The Pachycormiformes were a successful group of stem-teleosts. Although they persisted for more than 100 million years in the Mesozoic seas and occupied a significant space of the ecosystem, little is known about this most diverse group. One of the earliest pachycormiformes is the lower Jurassic Pachycormus bollensis. A reconstruction of the phylogeny with the early P. bollensis together with more derived and earlier studied species has yielded important information about the relationships within the group. Both a parsimony analysis and a Bayesian analysis were performed. Three exceptionally complete specimens of P. bollensis from the Holzmaden-locality were used for the study. The resulting trees strongly supported pachycormiform monophyly. Three major ecomorphological clades were returned from the analyses: the filter-feeders, hyper carnivores and a more generalist predator radiation that included P. bollensis. Despite, node support within the pachycormiformes are generally weak. The tooth-structure and the phylogenetic position of P. bollensis might indicate an intermediate grade between the filter-feeders and the hyper carnivores. / Pachycormiformer var en mycket framgångsrik ordning utav stam-teleoster. Trots att de överlevde i mer än 100 miljoner år i de Mesosoiska haven och erövrat ett flertal olika platser i ekosystemet, är denna mågfacetterade grupp som helhet tämligen okänd. Pachycormus bollensis från yngre jura är en av de tidigste arterna i denna grupp. En fylogenetiskt rekonstruktion av den tidiga P. bollensis tillsammans med yngre, mer utvecklade arter kan ge viktig information om släktskapen inom gruppen. En parsimonisk analys och en Bayesian analys utfördes. Tre exceptionellt bevarade exemplar av Pachycormus bollensis från Holzmaden användes i studien. Resultatet visar på ett stark monofyletiskt sammanband inom pachycormiformerna. Båda analyserna visade robusta resultat för tre övergripande grupper; filtrerare, karnivorer och en mer generell predator linje som inkluderar P.  bollensis. Dock hade de interna släktskapen inom varje större grupp svagt stöd. Tand-stukturen tillsammans med den fylogenetiska placeringen av P. bollensis kan tyda på en intermediär grupp mellan filterarna och karnivorerna.

<p>An important problem in the analysis of gene expression data is the identification of groups of features that are coherently expressed. For example, one often wishes to know whether a group of genes, clustered because of correlation in one data set, is still highly co-expressed in another data set. For some microarray platforms there are many, relatively short, probes for each gene of interest. In this case, it is possible that a given probe is not measuring its targeted transcript, but rather a different gene with a similar region (called cross-hybridization). Similarly, the incorrect mapping of short nucleotide sequences to a target gene is a common issue related to the young technology producing RNA-Seq data. The expression pattern across samples is a valuable source of information, which can be used to address distinct problems through the application of factor models. Our first study is focused on the identification of the presence/absence status of a gene in a sample. We compare our factor model to state-of-the-art detection methods; the results suggest superior performance of the factor analysis for detecting transcripts. In the second study, we apply factor models to investigate gene modules (groups of coherently expressed genes). Variation in the number of copies of regions of the genome is a well known and important feature of most cancers. Copy number alteration is detected for a group of genes in breast cancer; our goal is to examine this abnormality in the same chromosomal region for other types of tumors (Ovarian, Lung and Brain). In the third application, the expression pattern related to RNA-Seq count data is evaluated through a factor model based on the Poisson distribution. Here, the presence/absence of coherent patterns is closely associated with the number of incorrect read mappings. The final study of this dissertation is dedicated to the analysis of multi-factor models with linear and non-linear structure of interactions between latent factors. The interaction terms can have important implications in the model; they represent relationships between genes which cannot be captured in an ordinary analysis.</p> / Dissertation

To assess the losses associated with future earthquakes, seismic vulnerability functions are commonly used to correlate the damage or loss of a structure to the level of seismic intensity. A common procedure in seismic vulnerability assessment is to estimate the seismic fragility, which is defined as the conditional probability that a structure fails to meet the specific performance level for given level of seismic intensity. This dissertation proposes a methodology to estimate the fragility of corroded reinforced concrete (RC) bridges with two-column bents subject to seismic excitation. Seismic fragility functions are first developed for the RC bridges with two-column bents. All available information from science/engineering laws, numerical analysis, laboratory experiments, and field measurements has been used to construct the proper form of the fragility functions. The fragility functions are formulated, at the individual column, bent, and bridge levels, in terms of the spectral acceleration and the ratio between the peak ground velocity and the peak ground acceleration. The developed fragility functions properly account for the prevailing uncertainties in fragility estimation. The probabilistic capacity and demand models are then combined with the probabilistic models for chloride-induced corrosion and the time-dependent corrosion rate. The fragility estimates for corroded RC bridges incorporates the uncertainties in the parameters of capacity and demand models, and the inexactness (or model error) in modeling the material deterioration, structural capacity, and seismic demands. The proposed methodology is illustrated by developing the fragility functions for an example RC bridge with 11 two-column bents representing current construction in California. The developed fragility functions provide valuable information to allocate and spend available funds for the design, maintenance, and retrofitting of structures and networks. This study regarding the vulnerability of corroding RC bridges will be of direct value to those making decisions about the condition assessment, residual life, and the ability of lifeline structures to withstand future seismic demands.

It remains unlikely that a terrorist organization could produce or procure an actual nuclear weapon. However, the construction of a radiological dispersal device (RDD) from commercially produced radioactive sources and conventional explosives could inflict moderate human casualties and significant economic damage. The vast availability of radioactive sources and the nearly limitless methods of dispersing them demand an inclusive study of the acquisition pathways for an RDD. A complete network depicting the possible acquisition pathways for an RDD could be subjected to predictive modeling in order to determine the most likely pathway an adversary might take. In this work, a comprehensive network of RDD acquisition pathways was developed and analyzed utilizing the Bayesian network analysis software, Netica. The network includes variable inputs and motivations that can be adjusted to model different adversaries. Also, the inclusion of evidence nodes facilitates the integration of real-time intelligence with RDD plot predictions. A sensitivity analysis was first performed to determine which nodes had the greatest impact on successful completion of RDD acquisition. These results detail which portions of the acquisition pathways are most vulnerable to law enforcement intervention. Next, a series of case studies was analyzed that modeled specific adversarial organizations. The analysis demonstrates various features of the constructed Bayesian RDD acquisition network and provides examples of how this tool can be utilized by intelligence analysts and law enforcement agencies. Finally, extreme cases were studied in which the adversary was given the maximum and minimum amount of resources in order to determine the limitations of this model. The aggregated results show that successful RDD acquisition is mostly dependent on the adversary’s resources. Furthermore, the network suggests that securing radiological materials has the greatest effect on interdicting possible RDD plots. Limitations of this work include a heavy dependence on conditional probabilities that were derived from intuition, as opposed to actual historical data which does not exist. However, the model can be updated as attempted or successful RDD plots emerge in the future. This work presents the first probabilistic model of RDD acquisition pathways that integrates adversary motivations and resources with evidence of specific RDD threats.

Due to the strong genetic component of dental development, research has shown that mineralization patterns of the human dentition are relatively buffered against environmental influences that normally affect bone growth and development. It is because of this resistance to environmental factors and the continuous growth of the permanent dentition throughout childhood and adolescence that the evaluation of dental development patterns has become the preferred method of age estimation in living and deceased children. Researchers (Harris and Mckee 1990; Tompkins 1996; Blankenship et al. 2007; Kasper et al. 2009) have suggested that the timing of dental development varies by ancestral descent and geographic populations. However, further evaluations of these perceived differences in the timing of dental development among populations are necessary as classical statistical methods result in age estimations that are biased toward the age structure of the reference population. However, the Bayesian approach is beneficial since it incorporates relevant prior knowledge into the analysis and formalizes the relationship between assumptions and conclusions (Buck et al. 1996). Therefore, the purpose of this research is to incorporate methods in Bayesian analysis to compare the timing of dental development between two contemporary populations of the Southeastern United States, as well as test the accuracy of dental development age parameters devised by Moorrees et al. (1963) on a contemporary Florida Population. For this study, 51 panoramic radiographs of individuals from a contemporary Florida population ranging in age from 7.7-20.4 years were reviewed. Statistical analyses incorporated a Bayesian approach to compare the timing of dental development for individuals comprising the contemporary Florida sample with the timing of dental development for a contemporary Middle Tennessee population by utilizing the age structure of the Middle Tennessee population as informed prior knowledge, otherwise referred to as an informed prior. Transition distributions for age, given stage of dental development, were also modeled for individuals comprising the contemporary Florida sample. The accurate observation and comparison of probability density distributions for age can serve as a noninvasive method for evaluating the probability of whether or not an unknown individual is a particular age, given the stage of dental development. Results of this research indicate that there is a consistent underestimation of age for individuals comprising the contemporary Florida population when the age structure of the Middle Tennessee population is utilized as an informed prior. Additionally, the results of this thesis indicate that there is a consistent underestimation of age when utilizing age parameters of Moorrees et al. (1963) for the estimation of age for individuals from a contemporary Florida population. By incorporating a Bayesian approach to compare two contemporary populations of the Southeastern United States, a comprehensive analysis of the relationship between age and stage of dental development can be achieved. Therefore, the results of this thesis support Bayesian analysis as an appropriate method of evaluating perceived differences in the timing of dental development between contemporary populations. Furthermore, the results of this research are beneficial to the field of forensic anthropology as the observation of advanced stages of molar development utilizing panoramic radiographs serves as a noninvasive method in estimating age for unknown juveniles and young adults, and can also assist courts within the United States in determining whether or not an individual is legally considered a minor or an adult.