Masters Theses
https://hdl.handle.net/10161/2493
2018-11-18T07:46:00ZTwo Applications of Summary Statistics: Integrating Information Across Genes and Confidence Intervals With Missing Data
https://hdl.handle.net/10161/17533
Two Applications of Summary Statistics: Integrating Information Across Genes and Confidence Intervals With Missing Data
Mandan, Arpita
<p>Gene set enrichment methods are useful for the mapping of individual genes or proteins to pathways and signatures. We use this approach to study the expression levels of proteins encoded by different genes, and compare individuals that have Alzheimerâ€™s disease (AD) to those that are cognitively normal (CN). Different gene sets might show differential enrichment in the two classes. A correlation statistic is computed for measuring the correlation of a sample to one class rather than to the other, with respect to a gene. This allows us to find the enrichment score for the sample with respect to an entire gene set, and to analyze the gene sets that are differentially expressed in the two classes. The linear model is a powerful tool that we use to estimate the correlation statistic, thus accounting for the class, and also the other covariates such as age and sex of the individual.</p><p>We study the Jeffreys and Clopper-Pearson intervals for binomial proportions when we have missing data. We use multiple imputation (MI) to deal with missing data. Using simulation studies, we compare the MI Wilson, MI Clopper-Pearson, and the MI Jefferys intervals. We then show that the MI Wilson interval has better repeated sampling properties among all in the case of high missingness. In the case of low missingness, the MI Wilson and MI Clopper-Pearson produce similar empirical coverage rates that are close to the nominal coverage. For a very low value of the binomial proportion, the Jeffreys interval has the largest coverage with the smallest average interval length.</p>
Master's thesis
2018-01-01T00:00:00ZBiotic Interactions in the Genus Anthurium Schott (Araceae)
https://hdl.handle.net/10161/17531
Biotic Interactions in the Genus Anthurium Schott (Araceae)
Hartley, Nathan
<p>The genus Anthurium represents one of the largest genera restricted to the neotropics and has long been recognized as one of the "megagenera" of flowering plants, in addition to claiming the bulk of diversity (~25%) in its family, the Araceae. Despite this vast and beguiling array of morphologic diversity, observations on the biotic interactions that Anthurium participate with other living organisms in are scant. Although these types of observations help establish foundational knowledge on the natural history of organisms and are well-known from other large clades of neotropical herbs (i.e., orchids), the few observations that have been made for Anthurium are scattered throughout the literature, and no attempt has been made to synthesize this information. Primary literature documents were procured largely through the Duke Univeristy Library system. A total of 316 discrete biotic interactions are referenced here, 22 of which are evidenced here as new. Previously considered to be pollinated primarily by euglossine bees (Dressler 1968, Williams & Dressler 1976; Croat 1980), this review evidences a diverse assemblage of biotic interactors that complements the taxonomic, morphologic, and ecologic diversity of Anthurium. It is hoped that this information can provide a starting point for current and future researchers to springboard from as they continue to demystify the evolutionary forces that facilitated the diversification of this understudied megagenus</p>
Master's thesis
2018-01-01T00:00:00ZImproving Stability and Selectivity in Electrochemical Reduction of Carbon Dioxide in an Aqueous Solution
https://hdl.handle.net/10161/17532
Improving Stability and Selectivity in Electrochemical Reduction of Carbon Dioxide in an Aqueous Solution
Ji, Dong
<p>With the rising level of CO2 in the atmosphere, methods capable of converting CO2 into useful fuels are urgently needed. The electrochemical CO2 reduction has gained significant interest recently due to its ability to use renewable energies. However, the poor stability of catalysts in electrochemical CO2 reduction limit its application in industry. Here we have developed a light-involving method to remove the surface carbonaceous species which are believed to poison the catalysts. By taking advantage of plasmonic properties of the copper catalyst, the stability of the catalysts has apparently improved.</p><p>Another problem in electrochemical CO2 reduction is the poor selectivity. One of the main reasons is the existence of the side reaction, hydrogen evolution reaction. Here we have developed a catalyst by dispersing atomic nickel on nitrogen-doped winged carbon nanotubes with the ability to suppress hydrogen evolution during CO2 reduction. The Faradaic Efficiency of CO reached 90% at -1.6 V vs. AgCl/Ag reference electrode while the efficiency of HER had been suppressed to less than 10% in the optimal reaction condition. By comparing with Ni NPs, the suppression of HER can be directly observed in LSV curve. It is suggested that this suppression may result from the lack of adjacent active sites for the Tafel mechanism in HER.</p>
Master's thesis
2018-01-01T00:00:00ZAerodynamic Optimization of Helicopter Rotors using a Harmonic Balance Lifting Surface Technique
https://hdl.handle.net/10161/17530
Aerodynamic Optimization of Helicopter Rotors using a Harmonic Balance Lifting Surface Technique
Tedesco, Matthew Braxton
<p>This thesis concerns the optimization of the aerodynamic performance of conventional helicopter rotors, given a set of design variables to control the rotor's pitching angle, twist and chord distributions. Two models are presented for use. The lifting line model is a vortex lattice model that uses assumptions on the size and shape of the blade to simplify the model, but is unable to account for unsteady and small aspect ratio effects. The lifting surface model removes these assumptions and allows for a wider variety of accurate solutions, at the cost of overall computational complexity. The lifting surface model is chosen for analysis, and then condensed using static condensation and harmonic balance. The final system is discretized and pertinent values of power, force, and moment calculated using Kelvin's theorem and the unsteady Bernoulli equation. This system is then optimized in one of two ways: using a direct linear solve if possible, or the open source package IPOPT where necessary. The results of single-point and multi-point optimization demonstrate for low speed forward flight, the lifting line model is sufficient for modeling purposes. As the speed of the rotor increases, more unsteady effects become prominent in the system, and therefore the lifting surface model becomes more necessary. When conducting a chord optimization on the rotor, hysteresis effects and local minima are calculated for the non-linear optimization. The global minima within the set of captured local minima can be found through simple data visualization, and the global minima is confirmed to have similar behavior to the results of lifting line; a large spike in induced power at a critical advance ratio, with a sharp decline in induced power as the rotor flies faster. Within the realm of practical forward flight speeds of a conventional rotor, smooth, continuous results are demonstrated.</p>
Master's thesis
2018-01-01T00:00:00Z