Stormwater runoff is a major environmental concern, particularly in urban environments. Trends in managing stormwater have evolved (and continue to evolve) from a quantity only approach into a sustainable approach, which integrates quantity, quality, the environment, and aesthetics. Best management practices (BMPs) and Low Impact Development (LID) are two well-documented techniques capable of managing to sustainable standards. There are a number of stormwater models available to design professionals today. However, there are few which integrate site-scale BMP/LID analysis in a simplified fashion. The purpose of this study is to determine if there is a demand in the design profession for simplified stormwater modeling tools to help designers make informed decisions about integrating BMP/LID strategies into site plans. A Web-based questionnaire was administered to a group of design professionals to determine their knowledge of BMPs and their technological needs and preferences in meeting stormwater goals and requirements.

Giardia lamblia is a eukaryotic parasite that causes diarrhea in humans worldwide. Diarrheal diseases cause stunting and mental retardation in children in developing nations, therefore it is important to understand the molecular epidemiology of G. lamblia. Compounding this, it is not clear if companion animals such as dogs contribute to infections in humans through zoonotic transmission. The genotypes of G. lamblia that have been found in humans are A1, A2 and B, while those in dogs have been on rare occasions all three human genotypes, but largely C and D, which have only been reported in dogs and appear to be species-specific. The molecular epidemiology of G. lamblia in humans and dogs was assessed in an endemic region of Lima, Peru. With one exception, dogs were found to harbor the C and D dog genotypes of G. lamblia. A single family dog was found to harbor a human genotype of G. lamblia. A2 and B genotypes of G. lamblia, but not A1, were found in humans in the endemic region. Previous literature reported that A2 and B typing within genotype tools were available, however the A2 samples from the endemic region could not be distinguished from one another through nucleotide polymorphism sequence analysis. A molecular typing technique was developed to type A2 samples. The extensive sequence analysis performed on two chromosomes of G. lamblia, yielded different phylogenetic tree groupings for the same samples. This lead to algorithmic analysis, which demonstrated a significantly high probability that meiotic recombination is occurring in the A2 samples of G. lamblia. As G. lamblia is largely believed to be asexual, the conclusion of doctoral research performed in this study yielded controversial, yet significant evidence that sex in G. lamblia A2 genotype samples is indeed occurring.

The logical framework LF and its metalogic Twelf can be used to encode and reason about a wide variety of logics, languages, and other deductive systems in a formal, machine-checkable way. Recent studies have shown that ML-like languages can profitably be extended with a notion of subtyping called refinement types. A refinement type discipline uses an extra layer of term classification above the usual type system to more accurately capture certain properties of terms. I propose that adding refinement types to LF is both useful and practical. To support the claim, I exhibit an extension of LF with refinement types called LFR,work out important details of itsmetatheory, delineate a practical algorithmfor refinement type reconstruction, andpresent several case studies that highlight the utility of refinement types for formalized mathematics. In the end I find that refinement types and LF are a match made in heaven: refinements enable many rich new modes of expression, and the simplicity of LF ensures that they come at a modest cost.

Viral sequence data are increasingly being used to estimate evolutionary and epidemiological parameters to understand the dynamics of viral diseases. This thesis focuses on developing novel and improved computational methods for high-throughput analysis of large viral sequence datasets. I have developed a novel computational pipeline, Pipelign, to detect potentially unrelated sequences from groups of viral sequences during sequence alignment. Pipelign detected a large number of unrelated and mis-annotated sequences from several viral sequence datasets collected from GenBank. I subsequently developed ANVIL, a machine learning-based recombination detection and subtyping framework for pathogen sequences. ANVIL's performance was benchmarked using two large HIV datasets collected from the Los Alamos HIV Sequence Database and the UK HIV Drug Resistance Database, as well as on simulated data. Finally, I present a computational pipeline named Phlow, for rapid phylodynamic inference of heterochronous pathogen sequence data. Phlow is implemented with specialised and published analysis tools to infer important phylodynamic parameters from large datasets. Phlow was run with three empirical viral datasets and their outputs were compared with published results. These results show that Phlow is suitable for high-throughput exploratory phylodynamic analysis of large viral datasets. When combined, these three novel computational tools offer a comprehensive system for large scale viral sequence analysis addressing three important aspects: 1) establishing accurate evolutionary history, 2) recombination detection and subtyping, and 3) inferring phylodynamic history from heterochronous sequence datasets.

No description available.

京都大学 / 新制・課程博士 / 博士(人間健康科学) / 甲第23827号 / 人健博第98号 / 新制||人健||7(附属図書館) / 京都大学大学院医学研究科人間健康科学系専攻 / (主査)教授 澤本 伸克, 教授 林 悠, 教授 武藤 学 / 学位規則第4条第1項該当 / Doctor of Human Health Sciences / Kyoto University / DFAM

Classical phase II trial designs assume a patient population with a homogeneous tumor type and yield an estimate of a stochastic probability of tumor response. Clinically, however, oncology is moving towards identifying patients who are likely to respond to therapy using tumor subtyping based upon predictive markers. Such designs are called targeted designs (Simon, 2004). For a given phase I1 trial predictive markers may be defined prospectively (on the basis of previous results) or identified retrospectively on the basis of analysis of responding and non-responding tumors. For the prospective case we propose two Phase I1 targeted designs in which a) the trial is powered to detect the presence of responding subtype(s) as identified either prospectively or retrospectively by predictive markers or b) the trial is powered to achieve a desired precision in the smallest subtype. Relevant parameters in such a design include the prevalence of the smallest subtype of interest, the hypothesized response rate within that subtype, the expected total response rate, and the targeted probabilities of type I and II errors (α and β). (The expected total response rate is needed for design a) but not for b)). Extensions of this design to simultaneous or sequential multiple subtyping and imperfect assays for predictive markers will also be considered. The Phase II targeted design could be formulated as a single stage or Simon two-stage design. For multiple subtyping corrections to the significance level will be considered. Sample size calculations for different scenarios will be presented. An implication of this approach is that Phase II trials based upon classical designs are too small. On the other hand, trials involving "reasonable" numbers of patients must target relatively high threshold response rates within tumor subtypes. For the retrospective case we will provide the power to detect desired rates in the subtypes and provide the sample sizes required to achieve desired power. Retrospective analysis has the advantages that the analysis can be "supervised" by grouping responding and non-responding tumors; and multiple hypotheses, including hypotheses not formulated at the time of trial design, can be tested.

Thesis (MSc Food Sc)--Stellenbosch University, 2011. / ENGLISH ABSTRACT: Listeria monocytogenes is a foodborne pathogen that has been isolated from a variety of food sources. It is the cause of the food-borne disease, listeriosis that shows symptoms such as meningitis, encephalitis and abortion. Different strains of L. monocytogenes exist and not all are thought to be pathogenic to humans. The aim of this study was to evaluate and compare conventional methods, culturing on selective (Oxford agar) and chromogenic (RAPID’L.mono agar) media, as well as speciesspecific and multiplex polymerase chain reaction (PCR) methods for the detection and identification of 94 L. monocytogenes isolates from various areas in an avocado processing facility, as well as the final product. To achieve a better understanding of the genetic diversity of the confirmed L. monocytogenes strains isolated from the avocado facility, two subtyping techniques, PCR-restriction fragment length polymorphism (PCR-RFLP) and pulsed-field gel electrophoresis (PFGE), were employed. All of the isolates were identified as Listeria species on both Oxford and RAPID’L.mono agar. On the RAPID’L.mono agar, 76 of the 94 isolates produced colonies typical of L. monocytogenes, with the remaining 18 showing colonies typical of L. innocua (n=13) and L. ivanovii (n=5). The species-specific PCR successfully amplified a 730 base pairs region of the hly gene of 80 of the 94 isolates. For the same 80 isolates the multiplex PCR successfully amplified 800, 517 and 238 base pair (bp) fragments of the inlA, inlC and inlJ genes, respectively. The remaining 14 isolates included the 13 isolates identified as L. innocua, as well as an isolate identified as L. monocytogenes on RAPID’L.mono. The results obtained on the Oxford agar showed a 100 % positive correlation when compared to the PCR results in identifying Listeria species, while the RAPID’L.mono had a 4 % false negative result in identifying L. monocytogenes compared to the PCR results. Sixty-four of the confirmed L. monocytogenes isolates were subtyped using PCRRFLP and PFGE. For the PCR-RFLP analysis, a 733 bp fragment of the inlA gene was successfully amplified for all of the isolates, followed by digestion with the restriction enzymes, AluI and Tsp509I. AluI produced three different banding patterns and Tsp509I produced two different banding patterns. Subtyping of the isolates using PFGE was carried out by macrorestriction of the genomic DNA with ApaI and AscI. The restriction fragments were resolved by PFGE and the fingerprints were classified into four clusters. In the combined analyses, cluster I contained forty-eight isolates (n=48), cluster II 1 isolate (n=1), cluster III fifteen isolates (n=15) and cluster IV 1 isolate (n=1). The PCR-RFLP results had a 98 % correlation with the PFGE results. The results of this study indicated inconsistencies between the results obtained by conventional and molecular detection methods for the identification of L. monocytogenes. Species-specific and multiplex PCR, however, proved useful to accurately detect and identify L. monocytogenes in a shorter period of time and could replace the use of conventional agar during identification. Both PCR-RFLP and PFGE proved useful in the subtyping of L. monocytogenes isolates with the PCR-RFLP being less expensive and results obtainable in a shorter period of time. / AFRIKAANSE OPSOMMING: Listeria monocytogenes is ‘n patogeen afkomstig van voedsel wat uit ‘n verskeidenheid voedselbronne geisoleer kan word. Dit is die oorsaak van die voedsel afkomstigde siekte, listeriosis met simptome soos harsingvliesontsteking, ensefalitis en aborsie. ‘n Verskeidenheid L. monocytogenes stamme bestaan, maar nie almal word as patogenies beskou nie. Die doel van hierdie studie was om konvensionele metodes, naamlik mikrobiologiese kweking op selektiewe (Oxford agar) en chromatografiese (RAPID’L.mono agar) media, sowel as spesies-spesifieke en multipleks polimerase ketting reaksie (PKR) metodes te evalueer en vergelyk vir die deteksie en identifikasie van 94 L. monocytogenes isolate geisoleer vanuit verskeie areas in ‘n avokado prosesseringsfasiliteit sowel as die finale produk. Om ‘n beter begrip van die genetiese diversiteit van die isolate wat as L. monocytogenes bevestig is te verkry, is twee subtiperingstegnieke, PKR-restriksiefragmentlengte polimorfisme (PKR-RFLP) en pulsveld jel-elektroforese (PVJE) toegepas. Beide Oxford en RAPID’L.mono agar het al die isolate as Listeria spesies geidentifiseer. Op die RAPID’L.mono agar het 76 van die 94 isolate kolonies tipies van L. monocytogenes gevorm, 13 kolonies was tipies van L. innocua (n=13) en vyf kolonies tipies van L. ivanovii (n=5). Die spesies-spesifieke PKR het ‘n 730 basis paar (bp) streek van die hly geen suksesvol geamplifiseer vir 80 van die 94 isolate. Die multipleks PKR het 800, 517 en 238 bp fragmente van die inlA, inlC and inlJ gene onderskeidelik, vir dieselfde 80 isolate suksesvol geamplifiseer. Die oorblywende 14 isolate het die 13 isolate wat as L. innocua geïdentifiseer is en die een isolaat wat as L. monocytogenes op RAPID’L.mono geïdentifiseer is ingesluit. Resultate verkry met die Oxford agar het 100 % ooreengestem met die PKR resultate vir die identifikasie van Listeria spesies. Die RAPID’L.mono het ‘n 4 % vals negatiewe resultaat gelewer in vergelyking met die PKR resultate. Vier-en-sestig van die bevestigde L. monocytogenes isolate is gesubtipeer deur PKR-RFLP en PVJE. Tydens die PKR-RFLP analise is ‘n 733 bp fragment van die inlA geen suksesvol geamplifiseer, gevolg deur vertering met die restriksie-ensieme, AluI and Tsp509I. AluI het drie verskillende bandpatrone opgelewer en Tsp509I twee verskillende bandpatrone. Subtipering deur PVJE is uitgevoer deur makro-restriksie van die genomiese DNA met ApaI en AscI. Die restriksie fragmente is geskei deur PVJE en die vingerafdrukke is in vier groepe geklassifiseer. Groep I het 48 isolate (n=48), groep II 1 isolaat (n=1), groep III 15 isolate (n=15) en groep IV 1 isolaat (n=1) gehad tydens die gekombineerde analise. Die PKR-RFLP resultate het 98 % ooreengestem met die van die PVJE. Die resultate van hierdie studie het teenstrydighede tussen die resultate van konvensionele en molekulêre deteksie metodes opgelewer vir die identifikasie van L. monocytogenes. Die spesies-spesifieke en multipleks PKR het egter beide goed te pas gekom vir die akkurate deteksie en identifikasie van L. monocytogenes en kan heel moontlik die gebruik van konvensionele agar tydens identifikasie vervang. Beide PKRRFLP en PVJE was nuttig vir die subtipering van L. monocytogenes isolate. PKR-RFLP is egter ‘n goedkoper tegniek en die resultate is in ‘n korter tydsperiode beskikbaar.

Made available in DSpace on 2014-06-12T16:00:02Z (GMT). No. of bitstreams: 2 arquivo5819_1.pdf: 1022780 bytes, checksum: 817e10825cb544dad97eed36627fdd51 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2011 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / Uma abordagem formal e crucial na especificação e desenvolvimento de sistemas complexos. Inspirado pela engenharia, o desenvolvimento de software deve preterir a abordagem empirica e seguir uma abordagem estruturada, formal, passível de repetição e prova face ao advento de sistemas mais complexos, paralelos e concorrentes. Este trabalho apresenta uma extensão conservativa de OhCircus, uma linguagem de especificação oncorrente, que integra CSP, Z, orientação a objetos e um calculo de re- finamento. Esta extensão suporta a definição de heranca de processo, onde uxo de controle, operações e componentes de estado em um superprocesso, podem ser reusados por seus subprocessos. Neste trabalho nos apresentamos a gramatica estendida de OhCir- cus, acompanhada por um conjunto de regras de tipos que lidam com as novas construções da linguagem. Nos apresentamos, em termos da Unifying Theories of Programming definida por Hoare e He, a semântica formal de heranca de processo e suas construções de suporte. A principal contribuição deste trabalho e um conjunto, formalmente provado, de leis algebricas que lidam com herança de processo. Nós também explanamos informalmente como essas leis podem contribuir para uma teoria de completude para OhCircus. Finalmente nossas leis são exercitadas atraves de um estudo de caso

Blastocystis est un protozoaire anaérobie trouvé dans le tube digestif de l’homme et de nombreux animaux. Il est à ce jour le parasite intestinal le plus fréquemment retrouvé dans les selles humaines. Dix-sept sous-types (ST1 à ST17) ont été décrits en se basant sur la comparaison des séquences du gène de l’ARNr 18S. L’infection à Blastocystis est associée à une variété de troubles gastro-intestinaux et plusieurs études suggèrent une corrélation entre la pathogénicité et le ST du parasite. Trois différents axes de recherche ont été développés. Le premier s’est focalisé sur la prévalence et la biodiversité génétique de ce parasite dans les populations humaines. Des études épidémiologiques ont été menées en France et au Liban mais aussi en Afrique en réalisant la première enquête au Sénégal. Le sous-typage des isolats a été réalisé par PCR en temps réel en ciblant un domaine du gène de l’ARNr 18S suivi d’un séquençage direct du produit de PCR. Au Liban, la prévalence de Blastocystis était de 20% dans la population globale avec une corrélation entre le ST1 et le développement de symptômes gastro-intestinaux. Dans le même pays, cette prévalence dépassait les 60% chez des patients symptomatiques et des écoliers. Au Sénégal, la prévalence observée est la plus importante jamais décrite pour ce parasite puisqu’elle atteignait 100% dans une population d’une centaine d’enfants vivant en milieu rural. Ces données soulignent l’impact socioéconomique de la blastocystose dans les pays en développement où les conditions sanitaires sont souvent précaires. En France, une prévalence importante de 18% a pourtant été observée dans une large étude épidémiologique englobant des patients présentant ou non des symptômes et suivis dans 11 hôpitaux répartis sur tout le territoire français. Le ST3 est prédominant suivi des STs 1, 2 et 4 comme dans une majorité de pays à travers le monde. Le deuxième axe s’est concentré sur l’identification des facteurs de risque de transmission de Blastocystis à l’homme. Le parasite a été recherché dans les selles de vaches et de patients ainsi que dans des échantillons d’eau consommée par l’homme et les animaux dans une région géographique limitée du Nord Liban. 30% des échantillons humains, 69% des échantillons d'eau et 80% des échantillons de bovins étaient positifs pour le parasite. Le ST3 était prédominant dans les échantillons humains et d’eau suivi des ST1, ST2 et ST4. Par contre, ST10 et ST14 étaient prédominants chez les bovins mais ces deux STs n’ont pas été retrouvés dans les autres types d’échantillons. Pour expliquer l'absence des ST10 et ST14 dans ces échantillons, une transmission de ces STs par contact direct entre les bovins et/ou l'absence de formes kystiques transmissibles pour ces STs ont été proposées. Ce parasite a aussi été recherché dans les selles de nombreux groupes d’animaux du zoo de La Palmyre en France. Nous avons montré que près de 40% des selles analysés étaient positives pour Blastocystis et identifié de nouveaux réservoirs d'infections pour l’homme chez les carnivores. La prévalence du parasite atteignait 60% chez les primates chez lesquels les ST1 à ST5 identifiés sont identiques à ceux observés chez l'homme confirmant la faible spécificité d’hôte de ces STs. Dans une autre étude, la prévalence de Blastocystis était de seulement 3,5% dans une population de chiens en France suggérant que cet animal n'est pas un hôte naturel de Blastocystis. Enfin, pour clarifier la pathogénicité de ce parasite, le troisième axe de mes travaux a souligné le caractère invasif de Blastocystis dans un cas de péritonite appendiculaire chez une fillette de 9 ans de retour du Maroc. Seul Blastocystis a été détecté dans les selles, l’appendice, le liquide péritonéal et le sac de Douglas de cette patiente. Une gastro-entérite s’est de plus déclarée simultanément chez 26 membres de la famille de l'enfant suggérant une épidémie qui pourrait trouver son origine dans la consommation commune d’une eau contaminée. / Blastocystis sp. is an anaerobic parasitic protozoa found in the digestive tract of humans and numerous animals. To date, it is the most common intestinal parasite found in human feces with worldwide distribution. Seventeen subtypes (ST1-ST17) have been described based on the comparison of SSU rRNA gene sequences. Blastocystis infection is associated with various gastrointestinal disorders and many studies suggest a correlation between Blastocystis STs and pathogenicity. My work was developed on three different topics. The first concerned the prevalence and the genetic biodiversity of the parasite in human populations. Epidemiological studies were conducted in France and Lebanon but also in Africa by performing the first survey of this parasite in Senegal. Subtyping of the isolates was performed by real-time PCR targeting a domain of the SSU rRNA gene followed by direct sequencing of the PCR product. In Lebanon, the prevalence of Blastocystis reached 20% in the general population and we demonstrated a correlation between ST1 infection and the presence of symptoms. In the same country, this prevalence was 60% in schoolchildren and patients presenting gastrointestinal symptoms. Strikingly, the prevalence of Blastocystis in a population of one hundred children living in a rural area reached 100% in Senegal and more than half of the infected children by the parasite presented gastrointestinal disorders. These latter studies highlighted the socioeconomic impact of blastocystosis in developing countries with poor hygiene sanitation. In France, a large-scale molecular epidemiological study was performed including patients presenting or not gastrointestinal symptoms. Stool samples were collected during winter and summer in 11 hospitals spread all over the French territory. We observed a high prevalence of Blastocystis in the french population with an average of 18.2% and the predominance of ST3 followed by ST1, ST4 and ST2 as in numerous countries. We also identified seasonal variations since the average prevalence of the parasite is 13.6% in winter and 23.1% in summer. The second topic focused on the identification of the risk factors of Blastocystis transmission to humans. We searched this parasite in bovid and human stools as well as in drinking water samples consumed by bovids and breeders in a limited geographic area of North-Lebanon. 30% of human samples, 69% of water samples and 80% of bovid samples were positive for the parasite. Interestingly ST3 is predominant in human and water samples followed by ST1, ST2 and ST4. ST10 and ST14 were predominant in bovid but both STs are lacking in human and water samples. To explain the lack of ST10 and ST14 in human and water samples, we suggested a transmission of these STs occurring through direct contact between bovid and / or the absence of transmissible cystic forms of these STs. Furthermore, this parasite was searched in the stools of numerous animal groups in the zoo of La Palmyre in France. We showed that nearly 40% of the analyzed stools were positive for Blastocystis and identified new reservoirs of human infections in carnivores. The prevalence of the parasite reached 60% in primates in which the identified ST1 to ST5 are identical to those observed in humans confirming the limited host specificity of these STs. In another study, we showed that the prevalence of Blastocystis was of only 3.5% in a population of one hundred dogs in France suggesting that this pet is not a natural host of Blastocystis. Finally, to clarify the pathogenicity of this parasite, the third topic highlighted the invasive character of Blastocystis observed in a case of appendicular peritonitis in a 9-year old girl returning from Morocco. Only Blastocystis was detected in stools, appendix, peritoneal liquid and Douglas pouch of the patient. Interestingly, simultaneous gastroenteritis occurred in 26 members of the child’s family suggested an outbreak with contaminated water as probable origin.