Doctoral Dissertations

Permanent URI for this collection

https://hdl.handle.net/10919/11041

Browse

Now showing 1 - 20 of 17488

Potential Uses of Data Envelopment Analysis for Big Data and Machine Learning Integration in Evaluation
Sen, Anuradha (Virginia Tech, 2025-01-10)
Data science and evaluation are two disciplines that could benefit enormously from successful integration. While data science can provide useful information in terms of statistics for any evaluation process, it as a subject can benefit from incorporating elements that investigate the value addition aspects addressed by evaluation and evaluative thinking. Evaluative thinking when incorporated with the analysis of big data sources can become a powerful tool in our hands that could be used in various sectors, for various purposes, to build equity and justice in society and make us better prepared to deal with exigencies. The new advancements in data science tools and approaches have attracted researchers from various disciplines, including evaluators. However social scientists and evaluators have been slow adopters of data science technologies compared to other fields. If evaluation practitioners and scholars do not pay attention, then the other fields will race ahead with different methods, models, tools, and approaches (Bamberger, Raftree, and Olazabal, 2016). The new and emerging data sources and analytical approaches can not only help the evaluation practitioners to solve problems of various dimensions, but also the integration of disciplines like data science with evaluation has the potential to benefit each other and as well as for the promotion of social good. Data Envelopment Analysis (DEA) is a nonparametric linear programming approach to measure the efficiency and productivity of decision-making units (DMUs), which can be used in both public and private sectors. This approach can enhance the decision-making process by utilizing new emerging data sources. DEA, after being introduced in the late 1970s, has been widely used in the management sector, and over the last few decades, it has gained popularity in various additional fields due to being a data-driven approach. As far back as 1986, DEA was the subject of an entire textit{New Directions for Evaluation} volume, but it has rarely been explored again in the field of evaluation since then. Especially given the increased availability and use of big data in evaluation, there is a need to revisit DEA and explore its potential applications and implications for evaluation. This study is threefold, where it first explores what the published literature reveals about the nature of DEA and its relationship to the core evaluation concepts of evaluative thinking, reasoning, and ethics, using a scoping review. Secondly, it demonstrates a proof of concept of how DEA could be applied and the implications for evaluative thinking and reasoning using big data. Lastly, this study seeks the perspectives of thought leaders on the application of data science technologies and DEA in evaluation.
Optimizing Corn and Cotton Performance with Adaptive Management Systems and Subsurface Drip Irrigation in the Mid-Atlantic USA
Arinaitwe, Unius (Virginia Tech, 2025-01-10)
Corn (Zea mays L.) and cotton (Gossypium hirsutum) are globally important crops for food, feed, fuel, and industrial feedstocks. In Eastern Virginia, achieving optimal yields is challenging due to unpredictable environmental conditions which impact overall crop growth and nutrient use efficiency. More data are needed on adaptative corn management strategies that focus on increasing nutrient use efficiencies and crop yields (i.e., 4R nutrient management, biostimulants, and in-season crop protection chemicals). With less than 5% of Virginia's corn and cotton fields irrigated, increased irrigation adoption could stabilize/increase crop production outcomes in this region. Three studies were conducted to evaluate these management strategies with the following objectives: 1. To compare standard farmer practices with the Adaptive Corn Management System (ACMS) using a treatment omission/addition approach. 2. To analyze subsurface drip irrigation (SDI) effects on corn grain yield under different seeding and nitrogen (N) application rates in drought-prone soils of Eastern Virginia. 3. To evaluate SDI strategies in cotton, assessing the effects of dripline spacing, plant growth regulator (PGR) rates, irrigation strategies, N rates, and variety on yield. The first study integrated irrigation, in-season nutrient supplementation (soil and foliar applied), foliar fungicides, and biostimulants to enhance corn yields. Field trials conducted across five Virginia locations (2022 to 2023) with irrigated and non-irrigated sites showed yield improvements with supplemental nutrients, biostimulants, and fungicides in 4 of 10 experiments. The yield increase resulting from irrigation in intensive and standard management strategies ranged from -3 to 61%, averaging 8.9 and 34% for intensive and standard management practices, respectively. The second study (2022-2024) evaluated six SDI management strategies, four seeding rates (59,280 to 103,740 plants ha-1), and four N application rates (133 to 333 kg N ha⁻¹). Main effects of irrigation, seeding, and N rates significantly impacted yields. Irrigation and N interactions were significant across years for grain yield. Corn grain yield was greater by 102% with irrigation in 2022 compared to only 13%, and 51% in 2023 and 2024. Averaged over the three years the 0.91 m dripline and 0.91 m with volumetric water content (VWC) sensors increased revenue by $985 and $885 ha-1, respectively, above non-irrigated. Grain yield increased up to a seeding rate of 88,920 plants ha-1 and N rates up to 267 kg N ha-1. The third study utilized two experim¬¬¬¬¬ents evaluating SDI management strategies in cotton from 2019 - 2021. Experiment 1 of the third study tested three irrigation systems with various dripline spacings (0.91 m, 1.82 m, non-irrigated), four PGR rates (0%, 100%, 150%, 200% of current Virginia recommendations, and four cotton varieties. Experiment 2 examined three irrigation strategies (irrigation, fertigation, and non-irrigated), three N rates (89, 133, 178 kg ha-1), three PGR rates (0, 100 and 200%), and two cotton varieties. Results from experiment 1 showed that dripline spacing significantly influenced lint yields in 2 of 3 years. The PGR application rates significantly influenced lint yield in 2021 growing season only. Lint yield varied by variety in 3 of 3 years of the study. The 1.82 m dripline and 100% PGR rate produced the highest economic gains of $158 and $162 ha-1 respectively above check. In Experiment 2, the lint yield varied by irrigation all three years, while PGR rates, N application rates, and variety each influenced lint yield in 2 of 3 years. The highest rates of lint yield increase were achieved at 133 kg N ha-1. Irrigation implementation was more effective in increasing corn grain yields than cotton lint yields during the six-year study period. Corn grain yields were increased on average 60% with SDI compared to non-irrigated treatments over the three-year study. Inputs for adaptative corn management systems were not consistent for increasing grain yields. Although various PGR rates were evaluated, current PGR recommendations for cotton are sufficient with the varieties evaluated in maximizing lint yields. The current N applications for Virginia were in-line with those of the current study which identified 133 kg N ha-1. These studies provide the first data for corn and cotton management with SDI in Virginia and the Mid-Atlantic USA.
A mathematical analysis of evolutionary rescue and niche construction
Longcamp, Alexander Bascom (Virginia Tech, 2025-01-10)
Laboratory experiments and field studies have shown that declining populations can avoid extinction by undergoing adaptation. This process, evolutionary rescue, entails a population approaching extinction until an adaptive mutation appears and subsequently establishes by escaping stochastic loss. While most models of evolutionary rescue emphasize mutations that allow organisms to persist in hostile environments, a less explored possibility---known as positive niche construction (hereafter niche construction)---involves mutants improving their fitness by modifying their environment. In Chapter 1 of this dissertation, I analyze a model of evolutionary rescue via a niche-constructing mutation. I show that the probability of rescue is highest under low-to-moderate rates of construction: some construction is needed to ensure that mutants proliferate quickly enough to avoid stochastic extinction; but because construction is costly, requiring time and energy to perform and develop, too much of it can lead to over-exploitation of the constructed habitats by the mutants' non-niche-constructing ancestors (hereafter residents). In Chapter 2, I then model a niche-constructing population that must undergo evolutionary rescue to withstand habitat exploitation by an invading species. I find that the same fecundity costs rendering constructors vulnerable to exploitation can help facilitate rescue from such exploitation by reducing the rate of construction and thus lowering the density of habitats available to invaders. The lower habitat density leads to slower invasion, which in turn buys constructors more time to mutate. Finally, in Chapter 3, I consider the possibility that invaders directly interact with the resident population instead of exploiting resident-constructed habitats. I show that a lower resident birth rate---whether it stems from a smaller resident density independent birth rate or stronger birth-limiting competition---can promote rescue by reducing variance in mutant fitness. Together, these findings suggest that lower reproductive success among members of a population can, under a range of conditions, improve the population's chances of evolutionary rescue.
Investigating Sex-Specific Responses in a Preclinical Model of Traumatic Brain Injury: Development of Chronic Depression-Like Behavior and Glutamatergic Protein Changes
Talty, Caiti-Erin Teresa (Virginia Tech, 2025-01-09)
Concussion is the most common form of brain injury, comprising over 80% of traumatic brain injuries (TBIs) occurring in the United States and around the world. While many individuals are able to fully recover in the weeks following a concussion, an estimated 50% of patients go on to suffer from persistent symptoms that may range from months to years in duration. Among the most common complaints of those with persistent symptoms is sadness or depression, and significantly elevated rates of suicide have been reported in this population. Females are more likely to develop persistent symptoms and have reported higher rates of neuropsychiatric symptoms than males following injury. Altered glutamatergic neurotransmission has been implicated as a possible cause of depression following concussion due to similarities in glutamatergic changes that occur following concussion and during depression, independently of brain injury. Excitotoxicity is known to occur following TBI, resulting in neuronal death, and dysfunction in the cells that survive. Glutamatergic dysfunction occurring in regions such as the hippocampus, prefrontal cortex or amygdala, may be a driver of depressed mood or major depressive disorder (MDD) in concussed individuals as these regions, among others, have previously been linked to MDD. There are currently no approved pharmacological treatments for TBI symptoms. Therefore, gaining insight into chronic pathophysiology underlying TBI symptoms, including depression, is essential to support the development of therapeutic approaches for patients. The glutamate system represents a promising avenue of investigation in the context of chronic TBI pathophysiology. Using a clinically-relevant rodent model of concussion, this work sought to elucidate chronic glutamatergic changes occurring in the brain in association with the development of depression-like behavior following injury. Delayed-onset deficits in social and self-care behaviors were observed in association with region-specific changes in N-methyl-D-aspartate (NMDA) receptor and glutamate transporter expression in injured male animals. Females responded differently to injury, showing disinhibition and compulsive behaviors in conjunction with upregulation of glutamatergic signaling proteins. Further, this work aimed to investigate chronic sex-specific responses to TBI. Direct comparisons of behavioral changes in injured males and females demonstrated differences in both the emergence and nature of behavioral deficits. Examinations of hippocampal subregions showed deeper specificity in expressional changes in glutamatergic markers with both region- and sex-specific alterations observed at a chronic time point. Additionally, proteomic analysis was employed to evaluate widespread protein-level changes in the injured frontal cortex, and results revealed significant dysregulation in pathways involved in excitatory neurotransmission and calcium signaling. However, the dysregulated proteins within these pathways differed in a sex-dependent manner, indicating a sexual dimorphism in chronic TBI pathophysiology. Potential drug targets were also identified for investigation in future studies. The fundamental work presented in these studies provides strong evidence of dynamic, sex-specific modifications in the glutamate system in association with chronic TBI deficits in a translational model, ultimately providing a foundation for future development of therapeutic options to improve the lives of patients suffering with persistent symptoms.
The Exploration and Development of Focused Ultrasound Extraction (FUSE) for the Rapid Release of DNA from Complex Tissue Matrices
Stettinius, Alexia Danielle (Virginia Tech, 2025-01-09)
Over the past two decades, molecular detection platforms have seen rapid advancement, reshaping the way we monitor the safety and security of our environment and human health. One of the key drivers of this transformation has been the development of simpler, faster, and more accessible nucleic acid amplification tests (NAATs) that have enabled point-of-contact (POC) DNA testing, delivering real-time results in resource-limited settings. Despite these advancements, the DNA sample preparation process is laborious, resource-intensive, and often requires hazardous chemicals, preventing the performance of DNA extraction at the POC and severely limiting the potential of POC NAATs. Thus, DNA sample preparation methods have been the primary bottleneck restricting the widespread use and applicability of POC NAATs. To overcome this bottleneck, focused ultrasound extraction (FUSE) was recently introduced as a novel DNA extraction method capable of rapidly releasing DNA from complex tissue matrices without labor-intensive techniques or strong chemicals. This technology utilizes high-pressure focused ultrasound pulses that disintegrate tissue and release DNA through the control of acoustic cavitation. An initial feasibility study demonstrated the potential of FUSE for simple biological tissues, but the use of FUSE for preparing complex tissues has not been explored previously. Understanding the potential of the FUSE technology with diverse sample types is essential for developing versatile DNA preparation methods that can effectively protect both the environment and human health. This dissertation investigates the performance of FUSE in complex tissue matrices and evaluates the utility of a miniaturized FUSE system for streamlined DNA sample preparation. Specifically, this work addresses (1) the feasibility of FUSE in robust sample types with physical and chemical complexities that hinder DNA release, (2) the optimization of FUSE pulsing parameters to enhance the time efficiency of FUSE processing and improve the quality of released DNA, and (3) the development of a compact, accessible device for the performance of FUSE DNA sample preparation in resource-limited settings. The completion of this work will introduce a novel DNA sample preparation method to enable the use of POC NAATs.
The Oil Weapon Moment: The 1973 Oil Embargo and its Impacts on U.S. Energy Politics
Atalla, Basil George (Virginia Tech, 2025-01-09)
This dissertation examines the impacts of the 1973 Arab petrostate oil embargo on U.S. energy politics. I argue that the embargo was the moment that transformed oil from a domestic and highly regulated commodity into a matter of national security and competitive geopolitics. While its likelihood was foreseen by the Nixon Administration, the embargo did exacerbate an existing energy crisis that was caused by pre-embargo federal energy policies. Following the embargo, a dominant narrative emerged that viewed dependence on foreign oil supplies as an existential threat that merited extraordinary government measures. The securitization of the energy crisis allowed the Nixon Administration to implement many of its pre-embargo energy policies, including the launch of a national energy program to bring the U.S. to energy self-sufficiency by 1980. The embargo was the trigger for the creation of new governmental entities, such as the Department of Energy and U.S. Central Command, that endure to this day. It also shaped the U.S.'s close relationship with Saudi Arabia as an essential oil supplier and a key ally in the Arab world. The dissertation contests the revisionist accounts that argue that the embargo was a non-event, arguing that its impacts on U.S. domestic and foreign policies are still tangible and relevant.
Learning without Expert Labels for Multimodal Data
Maruf, Md Abdullah Al (Virginia Tech, 2025-01-09)
While advancements in deep learning have been largely possible due to the availability of large-scale labeled datasets, obtaining labeled datasets at the required granularity is challenging in many real-world applications, especially in scientific domains, due to the costly and labor-intensive nature of generating annotations. Hence, there is a need to develop new paradigms for learning that do not rely on expert-labeled data and can work even with indirect supervision. Approaches for learning with indirect supervision include unsupervised learning, self-supervised learning, weakly supervised learning, few-shot learning, and knowledge distillation. This thesis addresses these opportunities in the context of multi-modal data through three main contributions. First, this thesis proposes a novel Distance-aware Negative Sampling method for self-supervised Graph Representation Learning (GRL) that learns node representations directly from the graph structure by maximizing separation between distant nodes and maximizing cohesion among nearby nodes. Second, this thesis introduces effective modifications to weakly supervised semantic segmentation (WS3) models, such as stochastic aggregation to saliency maps that improve the learning of pseudo-ground truths from class-level coarse-grained labels and address the limitations of class activation maps. Finally, this thesis evaluates whether pre-trained Vision-Language Models (VLMs) contain the necessary scientific knowledge to identify and reason about biological traits from scientific images. The zero-shot performance of 12 large VLMs is evaluated on a novel VLM4Bio dataset, along with the effects of prompting and reasoning hallucinations are explored.
A Single Cell and Spatial Transcriptomic Investigation of Traumatic Brain Injury: Novel insights into endothelial-derived Eph signaling
de Jager, Caroline Dana (Virginia Tech, 2025-01-09)
A staggering number of injury related disabilities and deaths are connected to traumatic brain injury (TBI) worldwide. Traumatic brain injury (TBI) involves an intricate and multifaceted cascade of events, starting with an initial mechanical impact followed by secondary injury brought on by numerous physiological changes that involve significant dysfunction at the cellular and molecular level. One major predictor of severe TBI outcome is the extent of blood-brain barrier (BBB) disruption, which under normal conditions prevents the passage of bacteria, neurotoxins, and macromolecules from entering the brain. Disruption of the BBB is linked to worse clinical outcomes in patients in both the acute, subacute, and chronic phases. However, the principal mechanisms responsible for regulating BBB permeability, where, and for how long that permeability occurs following TBI remains to be elucidated. Previous research has shown increased mRNA and protein expression of ephrin receptor A4 (EphA4), a well-established axon guidance molecule, within hours and days following TBI. This study is the first comprehensive investigation of the role of endothelial cell-specific EphA4 in TBI on regulating the BBB using advanced techniques like single-cell RNA and spatial transcriptomic sequencing, in addition to our newly established dual dye-labeling system. The central hypothesis is that endothelial cell-specific deletion of EphA4 enhances BBB integrity, characterized by changes in single cell gene expression consistent with improved barrier function, altered cellular metabolism, and reduced neuroinflammation within the BBB niche. This hypothesis will be tested by leveraging spatial sequencing to identify upregulation of genes associated with BBB stability and neuroprotection and utilizing a novel approach for assessing BBB permeability that addresses the limitations of traditional Evans Blue Dye (EBD) assays, including lack of spatial resolution, enabling precise analysis of molecular weight-dependent extravasation patterns.
The Academic Achievement of Music Students vs. Non-Music Students in a Suburban High School in the Commonwealth of Virginia
Snead, Nicholas LaMar (Virginia Tech, 2025-01-09)
Participation in music education has proved successful for student academic success and positive youth development (Ilari, 2023). Consequently, educators and policy makers seek to understand what draws students to music and what impact participation in school music has on their academic and personal lives. Some researchers have recommended that school divisions take a more defined look at the benefits of students participating in music. The purpose of this study was to determine what difference, if any, existed in student academic achievement as measured by average daily attendance, grade point average, and office discipline referrals by race, gender, and students with disabilities between those enrolled in a music program and those not enrolled in a music program over a 3-year period in a suburban high school in the Commonwealth of Virginia. This study used quantitative data with a nonexperimental descriptive design to address the research questions posed. The results of the study revealed that music students had higher academic achievement than non-music students as measured by average daily attendance, grade point average, and office discipline referrals by race, gender, and students with a disability over a 3-year period from 2021-2024. The results of this study are significant for education leaders who want to examine the impact of music participation in secondary school. This study also yielded important information regarding demographic subgroups and students with disabilities.
The Development of a Locomotion Interface for Virtual Reality Terrain Simulation
He, An Chi (Virginia Tech, 2025-01-09)
Virtual reality (VR) technology has amazed us with its capability of blurring the boundary between real and virtual; it tricks our minds into believing that what we experience is real. This technology is used for entertainment and various applications such as rehabilitation, immersive training, 3D design, etc. However, navigating the VR world remains a significant issue that has yet to be solved. In VR, we often rely on controllers or joysticks to traverse the virtual world. VR navigation with joysticks creates a discrepancy between the visual and bodily senses, which creates extra cognitive load and other problems, such as VR motion sickness, which limit the long-term use of VR applications and prevent it from being widely adopted. Walking is the most common task that we perform every day; it is the most intuitive way of navigation and is desirable to be implemented in VR. This dissertation details the development of a device that enables walking in VR and allows users to experience different terrains. Researchers and companies have pulled off endeavors to enable locomotion in VR; some even achieved commercial success, like the Omni One or KAT VR treadmill. Still, most devices are limited in simulating flat-ground walking without the capability of displaying any terrain, which is far from our living environments and restricts its usage. Our living surroundings feature stairs, bumps, and slopes that those devices can not render. Furthermore, VR can be desirable in hazardous reaction training, which is often situated in highly unstructured environments. Being able to simulate various terrains not only enhances VR immersion but also extends its usage for simulating multiple scenarios that could require extensive cost to construct. This document presents the design, build, control, and evaluation of a robotic locomotion interface that aims to display computer-generated terrains, allowing realistic lower-body engagement. The device features a novel, compact design that allows it to be available in space-constrained places like rehabilitation clinics or smaller labs. It uses design guidelines derived from motion capture data with dynamic simulation to align the robot and human workspace. After that, the system framework is addressed from both hardware and software aspects. This device features open-source Institute for Human and Machine Cognition (IHMC) software integrated with the Simple Open Source EtherCAT Master (SOEM) library to execute real-time EtherCAT communication. An admittance controller has been implemented to achieve smooth physical human-robot interaction (pHRI), governing the robot motion according to user input force. This work presents measurements to evaluate the system's performance. The document presents a CoM (center-of-mass) estimation algorithm that is based on LIP (linear inverted pendulum) model and ZMP (zero-moment point). And the estimation method is further validated through two applications: an initial framework of tele-locomotion and VR interaction. In the first case, it uses estimated CoM motion as a tracking reference for humanoid robots. The second application presents a framework that is able to display virtual terrain in the physical world.
Enabling Routine Chemical Composition and Volatility Distribution Measurements of Aerosols
Kumar, Purushottam (Virginia Tech, 2025-01-09)
Traditional online measurements of the chemical composition and other physicochemical properties (such as volatility and oxygenation) of particulate matter have relied on expensive and complex research-grade instrumentation based on mass spectrometry and/or chromatography. However, routine monitoring requires lower-cost alternatives that can be operated autonomously, and such tools are lacking. Routine monitoring of particulate matter, especially organic aerosol, relies instead on offline techniques such as filter collection that require significant operator effort. To address this gap, first, we built a new online semi-continuous aerosol chemical composition monitor, the "ChemSpot", that provides information on volatility-resolved organic carbon and degree of oxygenation along with sulfur content at relatively moderate costs. Autonomous operation of the ChemSpot instrument was demonstrated for four weeks alongside a mass spectrometer (an Aerosol Chemical Speciation Monitor, or ACSM), and the results of the comparison were encouraging. Mean absolute percentage errors (MAPE) were estimated to be 21% and 27% for aerosol organic carbon and equivalent sulfate (equivalent amount of sulfate for ChemSpot measured sulfur content). Chemspot-measured oxygen-to-carbon ratio (O:C) compared well with ACSM-measured O:C for moderate aerosol loadings. Second, we extended the capability of the ChemSpot instrument to provide volatility distributions of organic aerosols. A thermogram-based method was developed for the ChemSpot for volatility calibration and the calculation of volatility distributions. This work also highlighted the need for better observational constraints on vapor pressure values from structure-activity relationship based models. Finally, the ChemSpot was deployed at a biomass-burning experiment (Georgia Wildfire Simulation Experiment, G-WISE) to show the utility of this instrument in studying changes in volatility distributions of Biomass Burning Organic Aerosols (BBOA) produced from different biomass fuel types (samples from Blue Ridge and Coastal Plains eco-regions of the state of Georgia), different burn conditions (prescribed burning vs. wild burning) and simulated atmospheric aging. Significant changes in the volatility distributions of organic carbon were observed for the two biomass fuel types studied. Prescribed burning led to the formation of some higher volatility organic compounds in the aerosols compared to the wild burning case. A similar but more pronounced observation of the formation of higher volatility organics was observed after the simulated atmospheric aging of the BBOA samples. The formation of these higher volatility organics could be because of the presence of higher moisture content during the prescribed burning conditions. The successful completion of these objectives provides confidence that the ChemSpot could be a viable tool for long-term data collection of aerosol composition and volatility and in turn advancing aerosol science and helping policymakers devise strategies to curb air pollution.
Cracking the code of cathepsin S: Structural determinants of specificity switching
DeHority, Riley Ambrose (Virginia Tech, 2025-01-09)
Cathepsin S is a cysteine protease in the papain family that digests antigens as part of the adaptive immune response, activates receptors, and is associated with extracellular matrix degradation in autoimmune diseases and cancer. A comprehensive review of the literature revealed potential pH- and redox-dependent specificity switches in the proteolytic specificity of cathepsin S. These were investigated through the digestion of peptides across a variety of pH and redox conditions. These experiments confirmed both pH- and redox-dependent patterns of proteolytic specificity, with narrowed specificity in alkaline and oxidizing conditions. An analysis of publicly available structures of cathepsin S identified a lysine residue which descends into the S3 pocket of the active site above pH 7.0, acting as a pH-dependent gate. Energy minimization of crystal structures show disorder in the loops which make up the active site of the protein, which increases in disorder when the disulfide bonds on the surface of cathepsin S are reduced. This is explored as a potential mechanism for the redox-dependent specificity identified in the digest experiments. These specificity switches may contribute to pathological structural damage attributed to cathepsin S, as pH and redox dysregulation are features of several cathepsin S-associated diseases.
Discovery and development of novel antifungal agents for the treatment of Candida auris infections
Elgammal, Yehia Abdallah (Virginia Tech, 2025-01-08)
Fungal infections are one of the leading causes of death in humans, causing infections that range from mild superficial infections to severe, and life-threatening invasive infections that affect the bloodstream and vital organs. Invasive fungal infections have a high mortality rate, leading to approximately 1.5 million deaths annually. The most common pathogens responsible for these infections are Candida, Cryptococcus, and Aspergillus. Currently, treatment options for invasive fungal infections are limited to three main classes of antifungal drugs: azoles, polyenes, and echinocandins. The emergence of new fungal species, such as Candida auris, which displays high resistance and mortality rates (30-60%), has further complicated treatment efforts. Thus, there is a critical need for new therapeutic strategies to combat these life-threatening pathogens. C. auris isolates have demonstrated significant resistance, especially to azoles (fluconazole) and polyenes (amphotericin B, AmB). To address this, we screened approximately 2,600 FDA-approved drugs and clinical compounds to identify agents capable of inhibiting C. auris growth and enhancing or restoring the antifungal activity of existing antifungals. This screening revealed that HIV protease inhibitors, such as lopinavir, atazanavir, saquinavir and ritonavir, significantly enhanced the antifungal activity of azoles (fluconazole, voriconazole, itraconazole, and posaconazole) and polyene (AmB). Mechanistic studies showed that the HIV protease inhibitors inhibited the fungal efflux pump, and interfered with glucose utilization, leading to reduced ATP levels in C. auris. Moreover, HIV protease inhibitors, in combination with AmB, were able to inhibit the virulence factors of Candida species. Furthermore, HIV protease inhibitors, in combination with itraconazole or posaconazole, resulted in a significant reduction of the C. auris burden in mice kidneys. Two additional drugs, darapladib and rilapladib, phospholipase A2 inhibitors, were identified as potent inhibitors of C. auris. Darapladib and rilapladib demonstrated superior killing kinetics compared to itraconazole. Interestingly, C. auris did not develop any detectable resistance to both drugs at sub-inhibitory concentration over 16-passages. Mechanistic studies revealed that darapladib increased the plasma membrane permeability and caused DNA leakage, likely due to a direct interaction with ergosterol, as suggested by competition assays with exogenous ergosterol. We further validated the broad-spectrum, fungicidal, and potent activity of darapladib in combination with AmB, showing significant synergy against multidrug-resistant fungal pathogens, including C. auris, Aspergillus fumigatus and mucormycosis-associated species. Additionally, darapladib demonstrated a superior safety profile compared to AmB, exhibiting lower affinity for cholesterol, and no toxicity in combination with AmB to kidney cells. Finally, rilapladib (at 2× MIC) inhibited the C. auris burden by 1.5 Log10, whereas darapladib (at 2× MIC) achieved complete eradication of the C. auris burden in an in vivo C. elegans model.
Investigating Volcano-Tectonic Interactions in the Natron Rift, East Africa with Implications for Understanding Volcanic Eruptive Processes
Masungulwa, Ntambila Simon Daud (Virginia Tech, 2025-01-07)
An early phase continental rift is an emerging plate boundary where tectonic forces stretch and thin the continental lithosphere, shaping the Earth's surface. Continental breakup and its progression are typically driven by the interplay between repeated magmatic and tectonic activities, which have been explored through both tectonic and magma-assisted rifting models. Understanding volcano-tectonic interactions is key for evaluating the role of magmatic fluids in facilitating the initiation of continental breakup during early phase rifting. This study applies the magma-assisted rifting model to the Natron Rift and investigates volcano-tectonic interactions during early phases of continental breakup associated with observed changes in the volcanic plumbing system of the active volcano Ol Doinyo Lengai. The Natron Rift is a magma-rich rift in the southern segment of the Eastern Branch in northern Tanzania providing an ideal setting to explore the interactions between tectonic and magmatic processes in the early stages of rifting. To investigate tectonic and magmatic interactions, we began by characterizing the magmatic plumbing system of Ol Doinyo Lengai using Global Navigational Satellite System (GNSS) data from our TZVOLCANO network and Interferometric Synthetic Aperture Radar (InSAR) observations. We inverted the GNSS and InSAR data independently to identify potential deformation sources using the software dMODELS. We then conducted a joint inversion of both datasets and found results that were consistent with the independent inversions within 2-sigma uncertainty. Our findings suggest that Ol Doinyo Lengai is fed by an offset multi-tiered reservoir system, consisting of a shallow magma reservoir located east of the volcano connected to a deeper reservoir through a network of fractures. This magmatic system likely influences the nature, style, and magnitude of volcanic activity at the edifice. We also assessed temporal and spatial changes in surface motion observed with GNSS stations associated with magmatic activity to help mitigate risks to nearby communities, tourism, and air traffic. Detecting transient deformation is essential for forecasting eruptions since these signals often precede eruptive events. To detect transient signals using GNSS data from the TZVOLCANO network, we employed the Targeted Projection Operator (TPO) program which projects GNSS time-series data onto a target spatial pattern. We analyzed seven years of continuous GNSS data and divided the observations into three-year intervals. The TPO method detected rapid uplift between March 2022 and December 2022 followed by steady-state uplift through August 2023. The method also identified quiescent periods and non-eruptive inflation signals that enhance our understanding of the dynamic magma plumbing system of Ol Doinyo Lengai. When integrated with the TZVOLCANO network, which streams real-time GNSS data, this approach enables continuous monitoring and early detection of potential volcanic hazards. Ongoing monitoring is crucial for assessing volcanic risks and improving emergency response plans. Finally, we examined the role of interactions between tectonic and magmatic processes in the Natron Rift during the early stages of continental breakup, focusing on the evolution of the magma plumbing system beneath Ol Doinyo Lengai. Using the code PyLith, we developed a 3D model of the region. The modeling experiments test both homogeneous and heterogeneous medium, with and without topography to estimate surface deformation and stress changes on the Natron fault due to geodetically constrained magma source inflation and deflation. Our analysis focused on stress transfer from the magma sources to assess the likelihood of fault slip, considering the typical 0.1 MPa threshold for triggering slip in magmatic rift settings. Results indicate that during the inflation period from 2016 to 2023, slip on the Natron fault is inhibited adjacent to the volcano under all scenarios. During the magma source deflation phase that occurred from 2007 to 2008 due to explosive eruptions, slip on the Natron fault was promoted adjacent to the volcano under all scenarios. Shear stress change analyses reveal that during the magma deflation scenario, slip of the Natron fault is consistent with oblique normal fault movement that is dominated by normal faulting and has components of strike-slip motion. Finite numerical modeling results demonstrate that topography considerably influences stress changes caused by dynamic magma sources as compared to material heterogeneity highlighting the importance of incorporating topography in volcano-tectonic settings. This work suggests that the potential ongoing magmatic activity at Ol Doinyo Lengai and its proximity to the Natron Fault influence the development of the youthful Natron Rift during early phase rifting. However, this influence likely inhibits fault slip at present on the adjacent section of the Natron fault due to magma source inflation.
Exploring the ecology of Orthobunyaviruses in Virginia and their pathogenesis in murine and poultry models
Lopez, Krisangel (Virginia Tech, 2025-01-07)
Cache Valley virus (CVV) is a vector-borne, negative-sense RNA virus, in the genus orthobunyavirus. Cache Valley Virus is a widespread pathogen in North America, and since its first isolation in 1956, has been associated with multiple epizootics of CVV in ruminants, leading to spontaneous abortions and congenital malformations. As such, CVV is a virus of high economic relevance, but little is known about fundamental aspects of its biology. To address this gap of knowledge, I conducted a series of studies to better understand the pathogenesis and ecology of CVV. This work is divided into two facets; the first is the development of animal models to assess the pathogenesis of CVV in various host species, and the second is vector surveillance to better understand the ecology of orthobunyaviruses within the Commonwealth of Virginia. In the first two chapters, I address the lack of small animal models to study CVV. First, I developed a murine model and an in utero model that mimic the natural progression of disease observed in CVV infection. In the second chapter, I study the growth kinetics of CVV in avian cell lines and in commercial poultry species. In the last chapter, I explore the distribution and diversity of mosquitoes and arthropod-borne viruses in Virginia. Overall, these studies provide insight into CVV pathogenesis and in utero transmission, the role of domestic poultry in the maintenance and amplification of CVV, and lastly, evidence of mosquito species range expansion, and high viral diversity across the Commonwealth of Virginia.
Exploring Radiomics and Unveiling Novel Qualitative Imaging Biomarkers for Glioma Diagnosis in Dogs
Garcia Mora, Josefa Karina (Virginia Tech, 2025-01-07)
Radiomics integrates machine learning (ML) and radiology to extract and analyze quantitative features from medical imaging modalities such as Magnetic Resonance Imaging (MRI), Computed Tomography (CT), Positron Emission Tomography (PET), ultrasound (US) and digital radiographs (DX). By extracting pixel/voxel-level data, followed by standardization and feature selection, radiomics enables ML algorithms to assist in diagnosis and prognosis. While extensively researched in human medicine its application in veterinary medicine remains limited. Radiomics offers objective, data-driven insights, surpassing qualitative evaluations by revealing micromolecular disease features invisible to the human eye. Radiomics holds significant promise for diagnosing gliomas (GM), a challenging brain tumor where histopathology, the diagnostic gold standard, is seldom performed in veterinary medicine due to logistical and financial barriers, and it is also limited by inherent pathologist subjectivity and disagreement. Additionally, qualitative MRI demonstrates limited accuracy in identifying GM type and grade. By offering non-invasive and reproducible diagnostic and prognostic solutions, radiomics has the potential to overcome these challenges, enhancing brain tumor evaluation in both veterinary and human medicine. The primary goal of this study is to enhance the diagnosis and prognosis of GM by exploring both conventional and innovative non-invasive imaging techniques, with a focus on qualitative and quantitative MRI approaches. We hypothesize that quantitative and novel qualitative methods will surpass conventional expert qualitative assessments in accurately diagnosing GM type, grade, and progression. By doing so, we aim to improve the precision of GM imaging diagnoses, offering clinicians a more accessible and reliable tool to support their diagnostic and treatment decisions. Chapter 1 of this dissertation presents a comprehensive review of the challenges associated with diagnosing GM using MRI. It also introduces principles of radiomics, a novel and relatively underexplored field in veterinary medicine centered on quantitative imaging analysis for diagnostic and prognostic purposes. This includes an in-depth discussion of the radiomics workflow and associated ML methods. Chapter 2 demonstrates the use and efficacy of quantitative MRI for determination of GM size and therapeutic response assessments using both linear and volumetric techniques. Chapter 3 investigates the T2-weighted–FLAIR mismatch sign (T2FMM) in dogs, a well-established imaging biomarker of human low-grade astrocytomas, and demonstrates that the T2FMM is a highly specific biomarker for oligodendrogliomas —the first such imaging biomarker for GM to be discovered in veterinary research. Finally, Chapter 4 illustrates a structured radiomics pipeline for the standardized quantitative analysis of brain tumors on MRI and demonstrates that the use of radiomics ML models results in superior ability to diagnose canine GM subtypes and grades and discriminate GM from non-neoplastic intra-axial lesions when compared to expert rater opinions derived from qualitative MRI evaluations.
Economic Benefits of Agricultural Conservation
Li, Yanggu (Virginia Tech, 2025-01-07)
Agricultural conservation provides a variety of public goods in the form of ecosystem services, such as improvement in water quality. The implementation of conservation practices may mitigate climate-related risks. The Environmental Quality Incentives Program (EQIP), a working agricultural land program in the United States, offers financial and technical support to farmers and ranchers who voluntarily adopt conservation practices. We estimate the economic benefits of EQIP in the rural housing markets of the US. The results suggest that the implementation of agricultural conservation projects increases local housing prices and that the improvement in environmental amenities is mostly capitalized by houses located within 25 miles. We study the effects of EQIP on agricultural production and land value. We find that a 10% increase in EQIP payments made in the previous two years is projected to increase corn yield by 0.03 bushel per acre at the county level in the US. A 10% increase in the previous year's EQIP payments for no-till practices is shown to increase wheat yield by 0.02 bushel per acre and decrease corn yield by 0.02 bushel per acre. We analyze the effects of conservation practices on acreage loss and farm loss. The results indicate that climate-smart payments have statistically significant effects on loss acres and loss cost ratios, such that a 10% increase in the previous year's and previous two years' climate-smart payments would reduce the loss acres of corn by four and seven acres, respectively. Moreover, a 10% increase in the previous year's climate-smart payments would reduce the loss cost ratios of corn by 0.02 percentage point, and the loss cost ratios of soybean by 0.01 percentage point.
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda)
Vasquez Valverde, Luisa Fernanda (Virginia Tech, 2025-01-07)
Academic abstract The millipede order Polydesmida, or flat-backed millipedes, is the most species rich among the class Diplopoda. At present, the order does not have a phylogenetic systematic classification, and the relationships of the 29 families are unknown. Whole genome DNA sequencing technologies and bioinformatic tools for the analysis of systematic data have improved over the last couple of decades, and have facilitated analysis of phylogenetic relationships. Here I sequenced the whole genomes of 82 species of the order Polydesmida and assembled a set of 281 orthologous genes using aTRAM. These genes were then used to estimate a maximum likelihood phylogeny. The families Paradoxosomatidae, Sphaeriodesmidae, Rhachodesmdidae, Platyrhacidae, Xystodesmidae, Polydesmidae, and Dalodesmidae were recovered as monophyletic groups. Some other families such as Trichopolydesmidae, Chelodesmidae and Macrosternodesmidae now appear to be polyphyletic groups, and a more comprehensive study of their molecular and morphological characters is needed. Phylogenetic trees can be used as maps to trace the evolution of characters in a group. Using the estimated tree for Polydesmida, I mapped morphological and molecular characters to understand their evolution. Traditional taxonomy in millipedes primarily used the morphology of the male gonopods to differentiate species. In my phylogeny, the morphological characters associated with the gonopod were less variable between specimens of the same family than non-gonopodal morphological characters, which showed a great diversity between individuals of the same family. A set of molecular characters that I studied were the gene order of the mitochondrion, as they may be a source of important evolutionary information. I found an inversion of half of the mitochondrial genome, which was consistent for the entire order Polydesmida. The majority of the changes in gene order were observed in the tRNA's, some of which can be considered synapomorphies for some families. Finally, I described three new species of millipedes in the superfamily Platyrhacoidea from southwestern Colombia: Aphelidesmus carcharodus and Pycnotropis svanae in the family Aphelidesmidae, and Barydesmus peineta in the family Platyrhacidae. I present diagnoses, descriptions, color habitus photographs and illustrations for these new species which contribute to the knowledge of millipedes of Colombia.
Identification and characterization of host genes involved in regulating replication of brome mosaic virus
Suseendran, Parkeswaran (Virginia Tech, 2025-01-07)
Brome mosaic virus (BMV) belongs to a viral class called positive-strand RNA [(+)RNA] viruses. This is the largest class of viruses and includes numerous important pathogens. BMV infects monocotyledonous plants and its replication can be recapitulated in the baker's yeast (Saccharomyces cerevisiae) under laboratory conditions to use yeast as an experimental model system. BMV generally does not infect dicotyledonous plants including the model plant Arabidopsis thaliana. One important shared feature of (+)RNA viruses is that they all make use of host proteins to aid in their own viral replication. In particular, (+)RNA viruses use host intracellular membranes for their replication and lipid composition of these membranes is crucial for viral replication. I show here that BMV replication protein 1a causes redistribution of host Lam5 (Lipid transfer protein Anchored at a Membrane contact site 5) and that Lam5 is necessary for BMV replication in yeast. Furthermore, in the absence of Lam5, BMV 1a affects the distribution of lipid droplets throughout yeast cells. Host factors also play critical roles in defense against viruses. Although wild-type Arabidopsis is not a host for BMV, the Arabidopsis cpr5 (Constitutive expression of Pathogenesis-Related genes 5) mutant can support systemic infection of BMV. I performed screens in Arabidopsis and have identified four genes that contribute to defense against BMV. These include two RNA-binding proteins, a lectin superfamily protein, and an alternative oxidase. My results also contribute to the growing evidence that reactive oxygen species play a key role in BMV replication. In summary, my work provided new insights into BMV replication in hosts and plant defense against BMV infection. The information gained from these projects aids in our understanding of (+)RNA virus biology in general and may identify targets for developing broad-spectrum antiviral strategies.
Bulk Ceramic-Based Biologically Inspired Composites: Design, Fabrication and Testing
Khan, Shahbaz Mahmood (Virginia Tech, 2025-01-06)
Strength and toughness are mutually exclusive mechanical properties; an increase in one result in the decline in the other. Accordingly, ceramics with superior strength have a very low toughness; likewise, metals with similar density have relatively lower strength but higher toughness. However, biological systems design lightweight materials, circumventing this limitation of conventional materials, by aggregating various multiscale toughening mechanisms. In challenging habitats, organisms evolve to produce remarkable multifunctional material systems that improve their "fit" and "survivability". Unlike traditional materials, natural materials employ special arrangements of structural elements into cellular, gradient, fibrous, layered, or overlapped "architected composites". These natural material systems are "architected" to delocalize damage and prevent defect coalescence, to avoid catastrophic failure, even though they are mainly composed of brittle building blocks (>90 vol% mineral content). Consequently, the study of natural materials has attracted the attention of scientists as the benchmark for the development of new synthetic materials. With the advent of additive manufacturing technology, the design and assessment of architected composites with bio-inspired motifs have become increasingly feasible. In this dissertation, I use multi-step fabrication methods with additive manufacturing as a key step to produce and study different biologically inspired architectures. With control over the design parameters of the architectural features, an in-depth understanding of the organization is accomplished. The case studies are primarily focused on bulk composite material systems with multiple phases and motifs inspired by various biological material systems. This dissertation aims to reveal the structure-property relationships of these structural motifs and the trade-offs to the mechanical robustness due growth-related constraints. With the help of stereolithographic additive manufacturing technique and centrifugal infiltration, we propose a bio-inspired method for preparing ceramic-metal composites. The approach allowed for flexible design, scalability, and dimensional control of individual phases. The ceramic-metal composites were fabricated with structures simplified from the mollusk shell architectures, exhibiting specific strength up to 169% higher than the base metal. The crack growth toughness of up to 12.9 MPa m1/2 was recorded, with crack deflection at ceramic-metal interfaces. Additionally, using tomographic analysis we show that the high porosities of 9% and 15% for green and sintered 3D printed parts, if improved, could further enhance the strength and fracture toughness of these composites. The outer protective layer of a bivalve mollusk exoskeleton, called the prismatic layer, is composed of normally oriented prismatic building blocks separated by soft organic matrix. The growth of the prismatic layer is regulated by the thermodynamic boundary conditions of the habitat and is directed from the exterior to the interior of the shell. A consequence of growth is a graded structure with a fine side (higher grain count with smaller grain size) and a coarse side (higher grain count with smaller grain size), however, the presence of grading results in asymmetry. Using mechanical testing we reveal that the organisms' selection of fine side as the loading face is "not the most optimized arrangement for templating". In fact, opting for the coarse side over the fine side as the loading side simultaneously enhances mutually exclusive properties such as stiffness, strength, and energy absorption. We further show that the curved prism motifs in the proximal parts of the Ostrea edulis shells result in a significant reduction in mechanical robustness due to the growth-related restrictions arising from the simultaneous normal and lateral growth of shells. Moreover, we show that although the addition of a nacre-like backing layer reduces the effects of axial directional asymmetry, the resistance of the prismatic layer to initiate damage in a coarse side-loaded hybrid composite is superior to the fine side-loaded counterpart. This part of the research highlights the need for caution when directly mimicking structural designs found in biological systems. Biological material systems are typically multifunctional, tailored to specific habitats and organism-specific needs, and often constrained by growth requirements and economic limitations. The shells of the pteropods – pelagic gastropod species, are comprised of helical or as posited by certain researchers "S-shaped" aragonite mineral motifs. These helical motifs are remarkably close packed in an organic matrix without noticeable spaces. We develop a biological process mimicking image processing technique called the "Bottom-up Sectional Morphing" to model perfectly closed packed structures with control over the radius and pitch of the helical motifs. With the developed composites we attempt to characterize the effect of the helix radius of individual motifs on the global mechanical properties. With the help of compressive tests, we characterize the delocalization of load as the radius of the helical motifs is increased. With the help of slab-shaped samples, we study the puncture resistance and interlocking behaviors due to increased helical radius. Using standardized fracture toughness tests, the toughness of the composites is determined. Additionally, the R-curve behaviors as a function of helical radiuses is characterized. On average, the fracture strength of the composite doubled as the radius of the helical motifs increased from 0 mm to 3.9 mm. Remarkably, the fracture toughness of helical composites was as high as 12-times the rule-of-mixtures estimated values. We summarize the extrinsic toughening mechanisms within the composites compared them to the mechanisms reported for helicoidal (twisted plywood) composites. Additional interlocking due to the uneven orientation of major axes in double basket weave pattern helical system are reported. Using explicit finite element simulations, we show that the curved motifs in comparison to normally oriented prisms, can help in developing localized high stress pockets, thus delocalization of damage that can help in increasing energy absorption during the progression of damage. Also, taking cues from fish scale ultrastructures, we design three-phase ceramic-epoxy-fiber composites. The fish scales feature gradient architectures with varying biomineralization extents from the distal to proximal regions (with respect to the fish body). From exterior to interior the mineralization content reduces, however, the collagen fiber count subsequently increases. To mimic the design approach, we use a 3D printed gradient ceramic lattice embedded in an epoxy matrix and backed using Kevlar fibers. With high-speed impact tests (73.5 ± 2.5 ms-1) we show that, although functionally graded composites (without Kevlar backing) show larger impact signatures compared to the similar density uniform density composites (without Kevlar backing) but absorb 35.7% higher energy during the process. High rebound velocity (22 ± 2.46 m/s) was observed for variable density composites with Kevlar backing. Additionally, using micro computed tomographic analysis of variable density composites with Kevlar backing we demonstrate that pre-stretching of fibers helps in the suppression crack. The results from this study were used in the design of polymer-elastomer composites with functionally graded material and fiber distribution. Interweaving fibers with hard solid lattices becomes challenging when one of the planar surfaces of the lattice is closed because of the functional grading. To overcome this challenge, I propose a new lattice interweaving method called "Warp-Assisted Binder-Tugging (WABT)", that can interweave the lattice using only one of the planar faces. Using WABT we refine the 3-phase composites design by incorporating strategically placed internal reinforcements. Cured photopolymer thermoset plastics are intrinsically brittle materials with mechanical properties like that of epoxy. Therefore, we choose this material along with urethane elastomer to prepare polymer-elastomer (hard-soft) composites, with and without reinforcements. We demonstrate the efficacy of strategic material distributions using dynamic puncture tests and projectile impact tests. The results show that concentrating brittle plastics towards the loading side improves energy absorption ability by 30.29% and puncture strength by 21.47%. A further 61.76% and 35.12% improvement in the energy absorption and puncture strength is recorded for slabs with backing and reinforcements. We show the response of the as-prepared composites under high speed projectile impact tests with incident projectile speeds of 151.5 ± 2.5 ms-1. The μ-CT characterization of damaged samples revealed the load delocalization and crack suppression behaviors due to the material distributions and reinforcements.

Browse

Recent Submissions