Electronic Thesis or Dissertation

Model
Digital Document
Publisher
Florida Atlantic University
Description
Topological Data Analysis (TDA) is a relatively new field of research that utilizes topological notions to extract discriminating features from data. Within TDA, persistent homology (PH) is a robust method to compute multi-dimensional geometric and topological features of a dataset. Because these features are often stable under certain perturbations of the underlying data, are often discriminating, and can be used for visualization of structure in high-dimensional data and in statistical and machine learning modeling, PH has attracted the interest of researchers across scientific disciplines and in many industry applications. However, computational costs may present challenges to effectively using PH in certain data contexts, and theoretical stability results may not hold in practice. In this dissertation, we develop an algorithm that can reduce the computation burden of computing persistent homology on point cloud data. Naming it Delaunay-Rips (DR), we define, implement, and empirically test this computationally tractable simplicial complex construction for computing persistent homology of Euclidean point cloud data. We demonstrate the practical robustness of DR for persistent homology in comparison with other simplical complexes in machine learning applications such as predicting sleep state from patient heart rate. To justify the theoretical stability of DR, we prove the stability of the Delaunay triangulation of a pointcloud P under perturbations of the points of P. Specifically, we impose a notion of genericity on the points of P to ensure stability. In the final chapter, we contribute to the field of computational biology by taking a data-driven approach to learn topological features of designed proteins from their persistence diagrams. We find correlations between the learned topological features and biochemical features to investigate how protein structure relates to features identified by subject-matter experts. We train several machine learning models to assess the performance of incorporating topological features into training with biochemical features. Using cover-tree differencing via entropy reduction (CDER), we identify distinguishing regions of the persistence diagrams of stable/unstable proteins. More notably, we find statistically significant improvement in classification performance (in terms of average precision score) for certain designed secondary structure topologies.
Model
Digital Document
Publisher
Florida Atlantic University
Description
Protein─protein interactions (PPIs) are essential for cell─cell interactions and cellular signal transduction, which play a crucial role in various human diseases. PPIs involved in cancer immunology pathways, known as immune checkpoints, have been intensely studied, leading to a new approach to cancer therapy. The PD1:PDL1 interaction is one of the most essential immune checkpoints. Studies on PD1:PDL1 showed over ten clinical monoclonal antibodies (mAb) used to treat cancer patients. Unfortunately, antibodies do not penetrate the tumor microenvironment well, and clearance from the body is slow, leading to unwanted side effects. There is a significant gap in the drug market between the typical Rule of 5 (Ro5) small-molecule drugs (MW<0.5 kDa, SASA ~150 Å) and large antibodies with molecular weights greater than 40 kDa (SASA >2,000 Å). PPIs remain challenging to modulate by small molecules due to their large, shallow, often dynamic, and water-exposed surfaces lacking well-defined binding pockets. Thus, our lab was drawn to work on large β-hairpin peptides (2-3 kDa) that can potentially mimic the CDR-H3 loops of some of the most potent and clinical anti-PD1 antibodies. Exploration of these β-hairpin peptides provided valuable insights into their folding stability, conformational flexibility, passive membrane permeability, and protein-protein interaction (PPI) blocking activities. Additionally, the rational design of TCIs against PD1, specifically targeting a lysine residue, emerged as a strategy to irreversibly obstruct the PD1:PDL1 protein-protein interaction enhancing potency of the non-covalent inhibitors by taking advantage of their specificity. Meticulous structural analysis, peptide synthesis, and biological evaluations are presented contributing comprehensions into covalent inhibitor development of drugs fbRo5.
Model
Digital Document
Publisher
Florida Atlantic University
Description
Plastic pollution in the marine environment is a global occurrence. Microplastics have been documented in numerous marine systems and organisms. Coastal estuaries and beach systems are at high risk for microplastic pollution. The distribution, abundance, and hazards microplastics present in these marine environments is not fully understood but are widely recognized as needed to support efforts aiming to protect and enhance these extremely valuable marine systems. This project aimed to quantify the abundance and variation of microplastics in estuarine mangrove and open coast beach sediments on Southeast Florida barrier islands, which are vulnerable and important coastal ecosystems. Barrier islands serve as a buffer between Florida’s wetland environments, reefs, and other marine habitats and may serve as a conduit or temporary sink for microplastics entering the ocean. The microplastic pollution present in estuarine mangrove and open coast beach systems may also elucidate patterns of microplastic pollution in the surrounding or similar coastal environments. There have been no extensive studies or monitoring efforts evaluating microplastics in Southeast Florida barrier islands sediments, nor comparing geomorphic properties of an area on microplastic accumulation. Study sites included back barrier estuarine mangroves and open coast beaches at three regionally similar but geomorphically distinct study sites throughout Palm Beach County, Florida. The sites were sampled seasonally in 2022 (i.e., summer and winter) to quantify the spatiotemporal distribution of microplastics.
Model
Digital Document
Publisher
Florida Atlantic University
Description
Social influence is sought to distribute information processing for decision-making when data is limited. Undergraduate students selected information with normative or informational wording to supplement a fabricated academic integrity appeal from their university and decided whether to affirm the charge. A novel measure, the Adaptive Scale of Preference for Normative Versus Informational Social Influence (ASPNVISI), was piloted in comparison with a Polish scale of influence preference, individual difference measures in motivation (e.g., Need for Closure), and behavioral measures of influence-seeking. Results did not support the hypotheses that psychological needs would predict behavioral social influence preferences, though Need for Cognition and Need to Belong predicted self-reported preferences. The ASPNVISI was correlated with the existing scale of influence preference, providing support for its continued development. Contrary to the hypothesis, confidence in the decision on a charge of academic dishonesty was not related to selected influence; race and conservatism were related to confidence, and gender to the selections. Next steps include further pilot testing of the ASPNVISI and expansion of the behavioral task.
Model
Digital Document
Publisher
Florida Atlantic University
Description
With the escalating challenges posed by global warming, sea-level rise, and natural disasters like hurricanes and tropical storms, coastal erosion has become a critical issue along the US coasts. The economic significance of U.S. coastlines, multitude of services offered by these coastal areas, underscores the critical importance of addressing the threats posed by both natural and human-induced factors that lead to erosion and coastal loss. To enhance future planning and to promote resilience of these invaluable coastal resources, it is essential to gain a comprehensive understanding of the strategies employed to mitigate coastal erosion in response to the diverse array of driving forces. A widely embraced solution to this erosion, caused by both natural forces and human activities, is beach nourishment.
Historical assessments of beach nourishment at the regional level have been conducted, however, there remains a gap in national-level analysis examining the overarching trends and the diverse factors that impact these trends. This study aims to provide a comprehensive perspective on beach nourishment practices encompassing 16 coastal states, including the top ten highly nourished states, and an additional six states selected from various coastal regions. It delves into the multifaceted factors that shape these practices, offering a holistic understanding of the beach nourishment landscape at a national level. An extensive analysis of beach nourishment trends and the influence of factors such as sea level rise, storms, hurricanes, hurricane categories, and coastal management approval years on beach nourishment activities was conducted.
Model
Digital Document
Publisher
Florida Atlantic University
Description
Demographics of the United States’ school system continue to change and diversify. Meeting the needs of diverse learners is critical to the success of all children. Building and sustaining equitable learning environments has proven to be a challenging task for school leaders. Barriers that hinder equity continue to plague schools and are rooted in systems of organizational injustice. Research suggests that leaders who are able to utilize socially just and culturally responsive leadership practices build and sustain equitable learning environments. School leaders have a large impact on the equity driven practices that exist within the organization; however, research indicates that they may not be prepared to lead such practices. Therefore, supporting the path of aspiring leaders through coaching, since they will eventually lead the charge, has great potential to be critical in the success of equity in our schools, and was the focus of this study.
The purpose of this qualitative case study was to explore and describe the coaching relationship between the principal and aspiring leader in order to understand how this relationship furthers the development of the aspiring leader into a school leader devoted to building and sustaining equitable learning environments.
Model
Digital Document
Publisher
Florida Atlantic University
Description
In this dissertation, I examined the lived experiences of people experiencing homelessness (PEH) and their perceptions of protective actions when making evacuation decisions during disasters or extreme weather events. Compared to other members of society, individuals experiencing homelessness during disasters face unique challenges, including a lack of shelter, transportation, access to news, and the need to carry all their belongings. This vulnerability is further heightened when natural disasters occur, as they lack physical protection, leading to increased exposure to hazards and adverse weather effects.
The literature on homeless evacuation decision-making is limited (Donley & Wright, 2012), and the direct experiences of PEH and their decision-making processes during natural disasters remain underexplored (Brookfield & Fitzgerald, 2018; Brown et al., 2013; Every & Thompson, 2014; Gaillard et al., 2019; Settembrino, 2017). Additionally, the unsafe living conditions of PEH, coupled with the uncertainty of the COVID-19 virus, posed unique challenges for homeless communities, increasing the risk of virus transmission and exacerbating social disparities.
In this study, I explored PEH’s lived experiences and perspectives when making evacuation decisions during disasters. I utilized Lindell and Perry's (2012) Protective Action Decision Model (PADM) as the theoretical framework. The PADM enabled me to analyze various factors influencing the evacuation decisions of PEH, including their physical aspects (shelter use and mobility) and social context (relationships with other PEH and employment). The findings presented in this study are the result of 41 in-depth semi-structured field interviews with PEH living unsheltered in Fort Lauderdale, Broward County.
Model
Digital Document
Publisher
Florida Atlantic University
Description
Misfolding and aggregation of Cellular Prion Protein (PrPc) is a major molecular process involved in the pathogenesis of Prion diseases. An N-terminal portion of the Prion protein, PrP106-128, is a 23-residue peptide fragment characterized by an amphipathic structure with two domains: a hydrophilic N-terminal domain and a hydrophobic C-terminal domain. Here, we studied the aggregation properties of the prion fragment peptide PrP106-128. The results show that the peptide aggregates in a concentration-dependent manner in an aqueous solution and that the aggregation is sensitive to pH and the preformed amyloid seeds.Furthermore, we show that the zwitterionic POPC liposomes moderately inhibit the aggregation of PrP(106–128), whereas POPC/cholesterol (8:2) vesicles facilitate peptide aggregation likely due to the increase of the lipid packing order and membrane rigidity in the presence of cholesterol. In addition, anionic lipid vesicles of POPG and POPG/cholesterol above a certain concentration accelerate the aggregation of the peptide remarkably. The strong electrostatic interactions between the N-terminal region of the peptide and POPG may constrain the conformational plasticity of the peptide, preventing insertion of the peptide into the inner side of the membrane and thus promoting fibrillation on the membrane surface. The results suggest that the charge properties of the membrane, the composition of the liposomes, and the rigidity of lipid packing are critical in determining peptide adsorption on the membrane surface and the efficiency of the membrane in catalyzing peptide oligomeric nucleation and amyloid formation.
Model
Digital Document
Publisher
Florida Atlantic University
Description
The representation of lead female characters as sexually threatening or naturally deceptive, hysterical, or evil, especially non-White or non-gender conforming characters, in popular science fiction and superhero film and television productions over the past few decades is concerning in that these films promote misogynistic and intersecting racist and hetero/sexist tropes in genres that typically appeal to younger audiences. Within their historical roots as cheap print entertainment, i.e., pulp magazines and comic books, directed at White working-class boys and young men, these genres have historically, and unabashedly, featured scantily clad, sometimes racially stereotyped, sexually titillating temptresses such as the Dragon Lady and Catwoman that threatened the hyper-masculine hero as well as humanity. Ignored by literary and cinematic critics throughout the twentieth century as juvenile male fantasy entertainment, the science fiction and superhero genres in film and television now dominate Hollywood productions.
Unfortunately, these genres in the twenty-first century still often promote damaging female tropes that suggest women as naturally defective, deceptive, power-hungry, irrational, raging monsters reminiscent of historical patriarchal myths of women. Additionally, a recent popular Netflix television series includes a character assigned female at birth (AFAB) who presents as gender non-conforming and carries attributes such as irrational rage and murderous violence that follows the historic cinematic trope of the “gleeful gay killer” as seen in Psycho (1960) and Dressed to Kill (1980). Although these themes in film and television are fantasy, they also mirror and bring to life the political and cultural anxieties of a significant number of men in our country who support the ideology of the manosphere that includes anti-feminist, anti-LGBTQ+, White supremacist, and racist beliefs. This dissertation examines three popular Hollywood films and one Emmy Award winning Netflix television series from the science-fiction and superhero genres since 1996 that reveal damaging female tropes that still prevail in popular entertainment.
Model
Digital Document
Publisher
Florida Atlantic University
Description
Gopher Tortoises are an imperiled terrestrial tortoise considered ecosystem engineers, as their burrows support over 350 species of “commensals”. Due to extensive anthropogenic habitat destruction and fragmentation, tortoise populations have significantly declined over the past 100 years. This isolation is leading to a premature loss of genetic variation, further precipitating the decline of this species. Understanding the genetic structure of this species can aid management decisions through intentionally relocating animals based on genetic units and historic gene flow, to preserve adaptive diversity and evolutionary potential on a large scale. I investigated whether Southeast Florida has a genetically distinct subpopulation of Gopher Tortoises in the unexamined southeasternmost extent of their range, from Palm Beach to Miami-Dade counties, using 9 microsatellite loci previously characterized for this species (Schwartz et al 2003) and found that our populations are only modestly differentiated from each other, and have greater genetic diversity than other populations.