eCommons Collection:http://hdl.handle.net/1813/472014-09-30T15:57:17Z2014-09-30T15:57:17ZCharacterization Of Human Lysophospholipid Acyltransferases In The Regulation Of Membrane TraffickingClarke, Benjaminhttp://hdl.handle.net/1813/371572014-08-19T18:35:00Z2014-05-25T00:00:00ZTitle: Characterization Of Human Lysophospholipid Acyltransferases In The Regulation Of Membrane Trafficking
Authors: Clarke, Benjamin
Abstract: Lysophospholipid acyltransferases (LPATs) catalyze the addition of an acyl chain to a lysophospholipid to form a phospholipid, dramatically altering lipid structure and behavior. These changes influence membrane curvature, the processes of vesicle and membrane tubule biogenesis, and the structure and trafficking dynamics of secretory organelles. At the beginning of my studies, I investigated the synergistic and antagonistic relationships between the activities of LPATs and phospholipases in regulating membrane trafficking. I went on to identify, by performing an overexpression screen, the human LPAT isoforms that are most important for regulating secretory membrane trafficking. I chose one of these enzymes for further characterization, the human membrane bound O-acyltransferase (MBOAT) gene family member lysophosphatidylcholine acyltransferase 3 (LPCAT3). Using selective membrane permeabilization immunostaining, I determined the topological orientation of LPCAT3's 11 transmembrane domains and luminal active site. I also found that a C-terminal K(x)KXX motif was necessary for LPCAT3 localization to the endoplasmic reticulum (ER). I observed the influence of LPCAT3 activity on the structure and dynamics of the early secretory system, through overexpression which reduced soluble protein secretion to 22% of control and caused relocalization of COPI and COPII vesicle markers. Additionally, LPCAT3 knockdown had profound effects on retrograde trafficking from the Golgi and ER-Golgi intermediate complex (ERGIC) to the ER. Knockdown slowed trafficking of the p58 receptor (ERGIC-53) and Brefeldin A induced recycling of Golgi phosphoprotein of 130 kDa to the ER. Slowed retrograde transport was accompanied by increased membrane tubulation. These effects of knockdown were reflected in the localization of ERGIC-53 and markers of COPII and COPI vesicles, which were mislocalized in knockdown cells. However, the alteration of LPCAT3 expression levels, through knockdown or overexpression, did not disrupt the morphology of the Golgi complex, trans-Golgi network (TGN), adaptor protein 1 (AP-1)/clathrin-coated vesicles, or endosomes. These results suggest that LPCAT3 activity is important for efficient COPI vesiculation, the production of retrograde membrane tubules, and the fusion of retrograde membrane tubules with the ER. These findings support a novel role for LPAT activity in the regulation of COPI function, membrane tubulation, and retrograde trafficking to the ER.2014-05-25T00:00:00ZAdvancing Electrochemical Energy Systems With Electron Microscopy ApproachesYu, Yingchaohttp://hdl.handle.net/1813/371502014-08-19T18:35:00Z2014-05-25T00:00:00ZTitle: Advancing Electrochemical Energy Systems With Electron Microscopy Approaches
Authors: Yu, Yingchao
Abstract: Achieving a secure and sustainable energy future is one of the greatest scientific, technological and societal challenges of our time. The projected doubling of world energy consumption within the next 50 years, along with our current dependence on nonrenewable fossil fuels, as well as the attendant detrimental effects on the environment that their utilization entails, point to the critical need of alternative energy sources as well as to the more efficient use of existing energy resources. The traditional fossil fuels such as coal, natural gas and oil rely heavily on the energy storage within the form of chemical bonds. A modern society must be able to store and convert large quantities of clean energy derived from solar, wind or other sustainable sources, as well as converting existing fossil fuels in a more efficient and sustainable way. This PhD thesis focuses on the use and development of advanced (scanning) transmission electron microscopy techniques, especially in-situ methods, for energy applications. Specifically, the intent is to understand materials properties and reaction mechanisms of fuel cell catalysts and novel lithium-ion battery and lithium-sulfur electrode materials, so as to enhance the performance of practical devices. Three topics will be discussed (i) advanced electron microscopy for electrochemical energy storage and conversion systems: from ex-situ to in-situ; (ii) catalysts for proton exchange membrane (PEM) fuel cells; and (iii) materials for advanced lithium-sulfur batteries.2014-05-25T00:00:00ZHow To Read The Saints: A Poetics Of Exemplarity In Sulpicius Severus' GallusYuzwa, Zacharyhttp://hdl.handle.net/1813/371522014-08-19T18:35:00Z2014-05-25T00:00:00ZTitle: How To Read The Saints: A Poetics Of Exemplarity In Sulpicius Severus' Gallus
Authors: Yuzwa, Zachary
Abstract: This dissertation argues that, in his writings on Martin, Sulpicius Severus constructs for his audience an ideal reader of hagiography, whose depiction allows him to condition the eventual reception of the text. Past scholarship on this corpus has focused especially on questions of historicity, in an attempt to understand more fully the figure of Martin in the context of a late ancient Gaul riven by ecclesiastical conflict. Instead of seeing Sulpicius' writing simply as a conduit to Martin, this project shifts scholarly focus from the holy man to his hagiographer. The dissertation's first chapter addresses the narrative structure of Sulpicius' writings on Martin, which include the original Life, three letters and the dialogue, Gallus. The second and third chapters account for Sulpicius' experimentation across diverse ancient genres: biography, epistolography and dialogue and argue that this formal progression allows Sulpicius to foreground the figure of the reader in the corpus. The chapters demonstrate that readers as depicted in the dialogue are marked as exemplary for Sulpicius' external audience: they model how to read a saint. The fourth chapter examines the content of that program of reading, in particular the frequent use of exempla in the dialogue, suggesting that Sulpicius uses these episodes to fashion a link that correlates the writing and reading of hagiography to the performance of saintly virtus.2014-05-25T00:00:00ZNanoporous Polymer/Ceramic Separator Electrolyte For Lithium Metal Battery ApplicationsTu, Zhengyuanhttp://hdl.handle.net/1813/371532014-08-19T18:35:00Z2014-05-25T00:00:00ZTitle: Nanoporous Polymer/Ceramic Separator Electrolyte For Lithium Metal Battery Applications
Authors: Tu, Zhengyuan2014-05-25T00:00:00ZComparing The Effect Of Downsizing And Reducing Pay On Collective Organizational Commitment And Firm Financial Performance And The Role Of Investment In TrainingYoon, Yeong Joonhttp://hdl.handle.net/1813/371512014-08-19T18:35:00Z2014-05-25T00:00:00ZTitle: Comparing The Effect Of Downsizing And Reducing Pay On Collective Organizational Commitment And Firm Financial Performance And The Role Of Investment In Training
Authors: Yoon, Yeong Joon
Abstract: This study compares the impact of downsizing and reducing pay on collective organizational commitment and firm financial performance. An examination of organizations operating under a salary schedule system in Korea shows that there is no significant difference between the effect of downsizing and reducing pay on collective organizational commitment and firm financial performance. However, compared to downsizing, reducing pay yields better financial performance for organizations that invest relatively highly in training of their employees through maintaining higher levels of organizational commitment. Keywords: Reducing pay, downsizing, pay reduction, investment in training, labor cost reduction strategy, organizational strategy, psychological contract, configurational approach2014-05-25T00:00:00ZHedging In Levy MarketsKirac, Yusufhttp://hdl.handle.net/1813/371492014-08-19T18:35:00Z2014-05-25T00:00:00ZTitle: Hedging In Levy Markets
Authors: Kirac, Yusuf
Abstract: This dissertation examines the hedging process in L´ vy markets. It consists of e three self-contained chapters. In Chapter 1, we study some results from Malliavin calculus and their applications to the hedging problem. We set up a market with finite number of assets each driven by a L´ vy process and focus on the hedging problem. A selfe financing trading strategy is designed to perfectly hedge (if possible) a given contingent claim. If a perfect hedging is not possible, it is designed to approximately hedge a given contingent claim. In general, such a market is not complete and only a small subset of square integrable contingent claims can be hedged. For other claims, a minimum variance hedging strategy is given in literature in terms of Malliavin derivative of contingent claims and assets in the market. We review this result. Such a finite market can be completed by power jump assets. Malliavin derivative of each power jump asset is a polynomial and these polynomials are dense in square Lebesgue-integrable functions. In particular, Malliavin derivative of a contingent claim gives us a square integrable function and it can be approximated by Malliavin derivative of power jump assets. This result shows that enlarged market by infinitely many power jump assets gives us a market complete in the limit. In other words, a contingent claim in such a market (complete in the limit) can be either hedged perfectly or there exists a sequence of contingent claims, each can be hedged perfectly, such that mean squared hedging error goes to zero. Chapter 2 focuses on hedging in Heath-Jarrow-Morton model (HJM). We assume that forward rates are driven by a L´ vy process and summarize no- e arbitrage conditions in HJM framework. We develop necessary and sufficient conditions for perfect hedging of a contingent claim by a self-financing trading strategy. These conditions give us two integral equations: one eliminates Brownian motion related risk and another eliminates jump risk. The latter one takes the form of a Fredholm integral equation of the first kind and it does not have a solution for all possible square integrable contingent claims. Both integral equations have kernels in terms of forward rates and both equations are solved for Malliavin derivatives in different directions. First, we derive the solutions for the hedging problem when the market is driven by a pure jump process. Our analysis shows that under certain conditions, all power jump assets can be perfectly hedged. This result gives us sufficient conditions for a market which is complete in the limit. Further, we extend our results for a general L´ vy process. e As an application of our approach, we develop a hedging strategy for a variance swap which can be approximately hedged. We also obtain the Malliavin derivative of a caplet. Chapter 3 investigates hedging in exponential L´ vy markets. First, we ree view the call option pricing model that is derived on the basis of Fourier transform. Using Malliavin calculus, we develop the necessary and sufficient conditions for perfect hedging of a contingent claim by a self-financing trading strategy of call options. These conditions give us two integral equations: the first one relates to hedge of Brownian motion related risk and the other deals with hedging jump risk. First, we assume that market is driven by a pure jump process and jump sizes are bounded. In this case, we have only one integral equation which is related to jump risk. The integral equation is transformed into a con- volution type integral equation and a closed form solution for a dense subset of square integrable contingent claims is obtained in terms of Fourier transform and Malliavin calculus. We extend these results to general L´ vy process with e bounded jump sizes. Our approach can be used in practice to develop dynamic hedging strategies using call options. The above results have important implications for hedging in exponential L´ vy market: one key implication is that for e each possible jump size we need a call option with a strike value equal to jump size. We also show some results related to Asian options in exponential L´ vy e markets. Finally, we outline the major limitations of our approach.2014-05-25T00:00:00ZCharacterization Of Single Defects In Zinc OxidePai, Yun-Yihttp://hdl.handle.net/1813/371472014-08-19T18:35:00Z2014-05-25T00:00:00ZTitle: Characterization Of Single Defects In Zinc Oxide
Authors: Pai, Yun-Yi
Abstract: This thesis summarizes my work on characterizing isolated defects in zinc oxide (ZnO). In chapter 1, I will briefly review the major strength of quantum information processing over classical computation. I will then review the basic properties of nitrogen-vacancy centers, the most-studied point defect species in diamond, and how they motivate the search for defects of similar properties in other semiconductor materials for defect-based quantum information processing. In chapter 2, I will describe our initial study on the optical properties of successfully isolated defects in ZnO: their fluorescence spectra, excited state lifetimes, and their photodynamics including blinking. In chapter 3, I will detail my work on extending the capability of an atomic force microscope to include simultaneous imaging in a confocal geometry. In chapter 4, I will discuss possible directions for our defect studies.2014-05-25T00:00:00ZThe Role Of Social Status In Negative Tie FormationLim, Yisookhttp://hdl.handle.net/1813/371452014-08-19T18:34:59Z2014-05-25T00:00:00ZTitle: The Role Of Social Status In Negative Tie Formation
Authors: Lim, Yisook
Abstract: This paper explores the question of how social status influences negative tie formation. Although previous literature in network scholarship has identified the role of social status in tie formation, it has mainly focused on the presence and the absence of positive ties such as friendship and exchange partnership as outcomes. However, another possible significant outcome has been neglected, namely the creation of negative ties. This relative neglect of negative ties has limited the understanding of status claims and network dynamics. In the current paper, I develop and test sets of competing possibilities about the role of status in the formation of negative ties. In using original 56 distinct social networks with negative tie data, I find that social status plays an important role in negative tie formation. In particular, I find that negative ties occur disproportionately from individuals of higher status and are directed towards individuals of lower status. The theoretical implications of these findings are discussed.2014-05-25T00:00:00ZReactivity And Continued Activity Of Immobilized Zinc Oxide Nanoparticles On Methyl Parathion DecontaminationHan, Yunfeihttp://hdl.handle.net/1813/371442014-08-19T18:34:59Z2014-05-25T00:00:00ZTitle: Reactivity And Continued Activity Of Immobilized Zinc Oxide Nanoparticles On Methyl Parathion Decontamination
Authors: Han, Yunfei2014-05-25T00:00:00ZIntelligible Models: Recovering Low Dimensional Additive Structure For Machine Learning ModelsLou, Yinhttp://hdl.handle.net/1813/371462014-08-19T18:35:00Z2014-05-25T00:00:00ZTitle: Intelligible Models: Recovering Low Dimensional Additive Structure For Machine Learning Models
Authors: Lou, Yin
Abstract: Different supervised learning models have different bias-variance tradeoffs. For low dimensional problems, low-bias models such boosted trees or SVMs with RBF kernels are very accurate but are unfortunately no longer interpretable by the users. For high dimensional problems, high-bias models such as regularized linear/logistic regressions are usually preferred over other models because of the curse of dimensionality and the exponentially growing hypothesis space but it is not clear whether we could further improve the accuracy from those high-bias models. Additive modeling is an excellent tool to control the bias and variance in a finer granularity and provides a great solution to these problems. Generalized additive models (GAMs) express the hypothesis as a sum of components, where each component can include any number of variables. Therefore, by prudently selecting the components or restricting the number of complex components and carefully controlling the complexity of each selected component, GAMs are very flexible of modeling hypothesis with different biases. This dissertation presents a family of additive models called intelligible models, which effectively recover the low dimensional additive structures. Those low dimensional additive components provide the opportunities for data scientists to investigate each simple component individually, and therefore the interpretability is significantly improved. We first present a large-scale empirical study of various methods for fitting GAMs. We demonstrate empirically that gradient boosting with shallow bagged trees yield the best accuracy. In ad- dition, we propose a very efficient method of detecting pairwise feature interactions that scales to thousands of features. With a large-scale empirical study, we show that models with low dimensional additive components (one- and twodimensional components) are as accurate as complex models such as random forests. Finally, we develop a method to carefully control the complexity of the intelligible models by feature selection and intelligently deciding whether the selected term is linear or nonlinear, and show that on high dimensional problems we can further improve the accuracy from the popular linear models by allowing a small set of features to act nonlinearly.2014-05-25T00:00:00Z