12466 research outputs found
Sort by
Computational Modeling of Francisella novicida Target Protein OppA with LL-37 Antimicrobial Peptide
Get PDFThis thesis has been embargoed for 2 years. It will not be available until November 2023 at the earliest.As a potential biological threat agent, the virulent bacterium Francisella tularensis remains an important topic of research. Understanding the mechanisms of action of antimicrobial peptides on F. tularensis is crucial to developing alternative treatments in the event of engineered or natural antibiotic resistance. The antimicrobial peptide LL-37 produces bactericidal effects in Francisella species and has demonstrated interaction with the membrane as well as periplasmic and intracellular proteins. One Francisella protein identified as an LL-37-binding protein is the periplasmic oligopeptide permease (Opp) complex protein, OppA. To perform computational structural modeling of the Francisella’s oligopeptide substrate-binding protein, crystal structure templates of homologs from several species were used to create models of F. novicida’s OppA sequence. To characterize the interaction of OppA and LL-37, computational docking was performed on the LL-37 fragment, KR-12 and then on the longer fragment LL-20. These studies revealed the favorable binding of KR-12 and LL-20 within the cavity of the OppA protein. This computational modeling supports the experimental data. These studies also revealed significant insight on the effect of open or closed protein conformations on the computational docking process. This approach illustrates the power of computational docking to reveal information about potential bacterial protein targets of antimicrobial peptides.2023-11-0
Optimization and Machine Learning Methods toward Improved Traffic Network Performance in Disrupted Environments
Get PDFMaintenance, traffic incidents, and other disruptions reduce roadway capacity, causing delays and safety concerns. Such events produce roadway downtime and it is crucial to understand its consequences. An ability to quantitatively understand such consequences is needed and can inform the development of mitigation strategies. Devising construction execution timing strategies, disseminating information to drivers during traffic disruption events, and proposing toll pricing changes for alternative tolled facilities during nonrecurring traffic conditions are such strategies considered here. This dissertation proposes mathematical models, algorithms and machine learning techniques to detect downtime events, assess their impacts on traffic and propose optimal remediation strategies. Contributions of this dissertation include: (1) a k-means clustering and local regression method for accurate identification of traffic event impact areas for understanding traffic event impacts and their extent over time and space; (2) a reverse engineering methodology using hybrid machine-learning methods with Support Vector Machines and Random Forests with k-Nearest Neighbor for detecting the occurrence of downtime events and their causes; (3) regression equations for estimating the traffic impacts of downtime events in terms of delays; (4) a multi-level, nonconvex mixed integer program for exploiting excess capacity of parallel tolled roadway facilities operated under public-private partnerships (P3) through optimal construction activity execution timing, reduced toll pricing and compensation to the P3 concessionaires; and (5) a dynamic systems model and analysis for determining optimal information dissemination strategies that promote stability in traffic flows and measure and enhance traffic system resilience to disruption. Together, these techniques enable improved understanding of traffic network disruption impacts and provide strategies for impact reduction and improved system resilience
Identifying the relationship between gut bacteria, host health and environmental factors in the red and maned wolf
Get PDFVarious factors threaten canid species globally, emphasizing the importance of zoo populations. However, zoo management is challenging, and individuals commonly display diseases that are not described in the wild. The red and the maned wolf are both impacted by gastrointestinal (GI) diseases from an unknown cause living under human care. This dissertation will i) address the link between fecal glucocorticoid metabolites (FGMs), gut bacteria and environmental factors in the red wolf, ii) identify relationship between FGMs and gut bacteria in the red wolf and iii) assess the relationship between gut bacteria, GI health and environmental factors in the maned wolf
EVALUATING THE RESILIENCE OF VISION TRANSFORMERS TO MEMBERSHIP INFERENCE ATTACKS
Get PDFMachine learning has come a long way in recent years. Adoption has begun to spreadand recent economic studies have shown the industry is expected to grow with billions of dollars being invested into the research and development of such technologies, which has been supported by the an economic impact study by The White House and the European Union [1]. Recently, the technology has seen a surge in use cases in domains which normally have not been active users of AI and Machine Learning. Specifically, medicine, finance, and national security/defence has picked up and starting taking advantage of AI and Machine Learning models [2] [3] [4]. The recent advances in Machine Learning architectures such as the transformer architecture has only sped up adoption and spread of the technology [5]. However, with the rapid innovation in the area, research needs to be done around security and privacy, to help build defences against attackers attempting to access sensitive data. Much research exists that has studied various attack methodologies against existing machine learning technologies, however, emerging innovative machine learning architectures have yet to be evaluated against existing, and potentially new types of attacks. In this work, we study the robustness of Vision Transformers (ViTs) to existing Member Inference attacks (MIAs), using a custom built, extendable, and salable framework. Vision Transformers are a novel proposal for adapting the Transformer Architecture to accomplish various vision related tasks such as classification [6]. Membership Inference Attacks are a type of adversarial attack in which the attacker attempts to gather information about the dataset on which a target model is trained on. This can lead to attackers accessing private data which may be sensitive. In our experimentation, we use our novel framework to experimentally explore the Vision Transformers robustness against different Membership Inference Attacks. Our experimental result, using the CIFAR-10 and CIFAR-100 datasets, show that ViTs models could provide a more reliable means of preserving the privacy of training data
Socially-Just Disability Resources: An Approach to Access and Equity in Higher Education
Get PDFThe increasingly pervasive special education teacher shortage has become a crisis in the United States, warranting immediate attention to minimize negative impacts on students with disabilities. More specifically, all efforts to mitigate the shortage by strengthening the special education teacher pipeline should be explored. One population of teacher candidates, teacher candidates with disabilities, would benefit from increased efforts from the teacher education field, as they consistently cite negative and inaccessible preparation experiences that hinder their ability to complete preparation programs necessary for teacher licensure. As a result, this dissertation will (a) investigate the experiences of teacher candidates with disabilities in teacher preparation programs through a systematic review of the literature from 2008 to 2022, and (b) explore an emergent professional paradigm in higher education disability resources as a potential means to enhancing access and equity in the experiences of teacher candidates with disabilities through a qualitative case study of one higher education disability resource center. Together, this dissertation aims to first present a key leverage point to stemming the shortage of special educators through attracting, preparing, and retaining educators with disabilities, and second, assess an approach for higher education disability resource professionals to lead the charge in enhancing the experiences of disabled teacher candidates. Implications for both higher education disability resources and special education teacher preparation programs will be presented
Characterization of Pullulanase (glgX/pulA) in Francisella novicida U112 Growth, Gene Expression, and Biofilm Formation
Get PDFFrancisella (F.) tularensis is a Gram-negative bacterium that causes the highly infectious human disease, tularemia. Within Francisella, like many other bacteria, polysaccharides play pivotal roles in many different pathways, whether they be intracellular polysaccharides or extracellular. One extracellular polysaccharide is pullulan; a water-soluble polysaccharide consisting of α(1,4) and α(1,6) glycosidic linkages and is cleaved by the enzyme pullulanase. Pullulan can be found everywhere in nature and is used in commercial food production. Interestingly, we noticed that Francisella encodes a gene for pullulanase. Within F. tularensis, pullulanase can be found within all four strains. In F. tularensis subsp. novicida (or F. novicida U112), the pullulanase gene is called glgX (Glycogen debranching enzyme), pulA and pulB depending on the annotation. Pullulanase (pul) is an enzyme that hydrolyze α(1,4) and α(1,6) glycosidic linkages. Within Francisella, the same genes annotated as type I and type II pullulanase can both be found. After a comparative analysis of the sequences, I conclude that Francisella pulA and pulB genes encode the same protein. My thesis focuses on characterizing the role of pullulanase in Francisella physiology and virulence. For my first aim, I identified the role pullulanase plays in Francisella in vitro growth stages (lag phase, exponential, and stationary phase) by characterizing the effect that a glgX mutant has under various conditions. Under this aim, I compared glgX mutants with WT F. novicida through various pH ranges, temperature ranges, and nutrient variations. I predicted that glgX will be a positive regulator for in vitro growth. Based on research conducted in K. oxytoca, E. coli, and Francisella on glgX-connected genes and biofilm regulators, I characterized the role pullulanase plays in biofilm formation in the second aim. Comparing all four strains of Francisella, F. novicida has the strongest evidence supporting the premise that biofilms are important for its environmental and stress stability. I compared WT F. novicida against glgX mutants in terms of biofilm morphology. I predicted that WT glgX is a positive regulator on biofilm production and mutating it will result in a negative regulation. A paper written by Uda et al demonstrated that in F. tularensis SCHU S9, pullulanase is an important factor in intracellular replication in macrophages but not in disease progression in mice. On the contrary, other literature reporting on the effect of mutants in glgX/pulA/pulB in Francisella suggests that pullulanase is not needed for intracellular replication. Overall, I expect to demonstrate that pullulanase (glgX) is important for in vitro growth, is a positive regulator in biofilm formation, and will have no effect on the intracellular replication of Francisella novicida U112
Surviving NIBRS: Restoring America’s Unreported Homicides and Exploring the Influences for Law Enforcement’s Declining Cooperation in Crime Reporting
Get PDFFBI adoption of the National Incident Based Reporting System (NIBRS) in 2021 as the mandatory reporting standard for crime data resulted in an unprecedented decline in police reporting to the federal government. Only 57 percent of the nation’s homicides were reported that year. This study obtained more than 6,000 unreported homicides from local and state police agencies using Freedom of Information Act and Open Record Act requests. The study compares FBI data and the study’s augmented dataset for accuracy and completeness using the National Vital Statistics System as a reference. This study also used a 3,134-county regression analysis to explore the socioeconomic factors associated with police decisions to participate, or to decline participation, in the more complex NIBRS program
Quarter 2 2023
Get PDFThis issue of GEWEX Quarterly contains the articles on the following: a commentary on GEWEX’s function and metrics for success; the new 2023 GEWEX Ambassadors; updates from YESS and the 2023–2024 Executive Committee (ExeCom) members and Regional Representatives; a new ESA Precursor Project on river discharge; summarizing the site-level state-of-the-art in TBM-SIF modeling, and ultimately advancing underlying equations and parameters to better match variations in fluorescence and photosynthesis across a range of environments, setting the stage for regional- to global-scale analysis and model-data integration with SIF-MIP; parameterizing sub-grid heterogeneous exchanges between the land and atmosphere and characterizing its implications for surface climate, variability, and extremes with CLASP; recent developments in the Asian Precipitation Experiment (AsiaPEX) and a proposal for the international field campaign Asian Monsoon Year II (AMY-II); and a report on regional water and energy budget closures and much more from the GEWEX Integrated Product Workshop
Reclaiming the Past, Demanding Futures: Queer Community in Kern County, California
Get PDFIn this thesis I use a mixed methods approach to investigate the ways in which queer communities in Kern County have survived systems that constrain their life possibilities, including the grip oil and carceral institutions have on Kern County’s economy and spatial imaginary. In contesting the myth of non-survivability in Kern County, I also contest the common perception that queer space and queer community does not exist outside urban spaces. Through archival work and oral histories, I recover some of the lost and suppressed memories of queer survival and queer community. In collaboration with local queer community members, I have co-created the Kern LGBTQ+ Community Archive, which is an online, free to access archive that houses the stories and histories compiled for this study
Three Essays on Economic Order and Intervention
Get PDFThis dissertation explores the nature of economics as a scientific discipline and the consequences of acting contrary to fundamental economic laws. Chapter One re-articulates Ludwig von Mises' argument concerning the categorical inability for planners to engage in economic calculation absent the institutional prerequisites of private property and exchange in goods of all orders, including higher order "means of production," by way of a common medium of exchange, in light of much-touted recent advances in "big data" and artificial intelligence (AI) technology. The major purpose of this chapter is to describe the most fundamental barrier limiting the ability of planners to utilize resources in an economic manner while operating outside the orbit of economic calculation, namely that ordinal preference rankings of consumers do not automatically generate commensurate cardinal units with which to engage in capital accounting. A further purpose is to demonstrate that no advances in computing power can overcome the need for the market pricing process in which individuals in their roles as both consumers and producers bid for economic goods and by doing so generate a unified structure of money prices. While technological advances such as "big data" and AI may be serviceable for entrepreneurs acting within a market setting characterized by prices assigned to all goods, to imagine that one day the problem of economic calculation under socialism can be overcome by such advances in computer technology is to misconstrue Mises' argument entirely and to confuse the science of economics with the science of technology. Chapter Two describes the economic approach to unintended consequences and synthesizes theoretical contributions on the topic by economists throughout the past three centuries. It then explores the empirical work on negative unintended consequences arising from government interventions in areas such as minimum wage laws, driver safety regulations, drug prohibition, intellectual property laws, and foreign military and humanitarian intervention. Although details vary across the cases, a uniting theme is that interventions often produce consequences at variance with their stated or assumed goals. However, due to the counterfactual nature of any theory dealing with causation, rebuttals might claim the situation would be even worse had the intervention not taken place. Therefore, it is vital to first establish correct a priori theoretical relations and then check if the relevant assumptions in theory correspond to the concrete conditions under investigation. Chapter Three examines "lessons learned" statements by U.S officials involved with the reconstruction effort following the U.S.-led war campaign in the early 2000s. The analysis is structured to reflect the fundamental epistemological problems facing foreign interveners at three distinct decision nodes. The first is the selection of a unified scale of ends to be pursued with respect to those for whose benefit the intervention is ostensibly occurring as well as within and across bureaucratic agencies. The second is the selection and implementation of the means by which the ends are to be pursued. The third concerns the realization of ends, whereby unacknowledged system effects generate outcomes unintended and possibly counterproductive from the point of view of the intervening party. Evidence from the Afghanistan Papers points to rampant corruption, the rise of warlordism, and a boom in narcotics production as the perhaps unintended but nevertheless real consequences of U.S. nation-building efforts in Afghanistan