12466 research outputs found
Sort by
PREDICTION OF SOLID TUMORS CHEMOTHERAPY EFFICACY USING MACHINE LEARNING APPLIED TO GENE EXPRESSION DATA
Full text linkSolid tumors present a significant challenge in oncology, often developing resistance to standard treatments. Recent advancements in cancer therapies have significantly improved patient survival rates and quality of life. Despite these advancements, treatment resistance frequently leads to tumor relapse and recurrence, contributing to a high number of cancer-related deaths. This resistance, caused by various mechanisms including changes in drug targets and activation of DNA repair mechanisms, remains a major obstacle in successful chemotherapy for solid tumors.The ability to predict how tumors will respond to specific treatments is crucial for developing personalized oncology strategies. Advances in molecular profiling and machine learning offer new opportunities to tackle this challenge by identifying gene signatures that predict treatment response with high accuracy. Our project aims to use machine learning to uncover these signatures in solid tumors, proposing a novel machine learning strategy to improve prediction accuracy and address overfitting. Our approach has shown promising results, with drug response prediction accuracy ranging between 85% and 96% across different types of solid tumors, including colorectal, serous ovarian, and triple-negative breast cancer. This success suggests that our method could, upon further validation, significantly impact clinical practice and precision oncology, paving the way for personalized treatment plans that could transform patient care
Biochemical Sequelae of Repetitive Mild Traumatic Brain Injury in a Murine Model Treated with Young Blood Plasma
Full text linkMild traumatic brain injuries (mTBI)—i.e., “concussions”—are experienced widely by athletes in contact sports and in military personnel. There is growing evidence to suggest that these populations are at increased risk to sustain repetitive mTBI (rmTBI) and that damage is cumulative with each subsequent injury. The pathology that succeeds rmTBI is linked to the development of other neurodegenerative disorders, such as Alzheimer’s disease (AD) and chronic traumatic encephalopathy (CTE). Within the last decade, researchers have discovered the promising effects of transfusing healthy young blood plasma for the reduction of neurological decline. Young blood plasma treatment has been shown to significantly ameliorate the neuronal damage, synaptic loss, and cognitive deficits associated with aging and AD. This study performed biochemical assays to investigate whether young blood plasma may also be an effective treatment for the neurodegenerative sequelae of rmTBI and CTE-type pathology in young mice. Brain and blood plasma samples were sourced from a previous doctoral dissertation that employed a controlled cortical impact model of rmTBI in C57Bl/6J mice starting at 8 weeks of age (Barkey, 2022). Half of the mice received one mTBI every 48 hours (5 mTBI total) while the other half received “sham” injuries as a healthy control. Following the rmTBI protocol, injury and sham mice received 150 µL healthy young blood plasma (donated from 8–10-week-old non-injury mice) or 150 µL saline every 48 hours for 16 total tail-vein injections. Plasma and saline injections were delivered in the form of either immediate treatment (24 hours after the last TBI) or delayed treatment (1 month after TBI). To examine the effects of immediate and delayed treatment following rmTBI, western blots measured brain lysate proteins related to memory function and synaptic plasticity—cAMP-response element binding (CREB) protein, phosphorylated CREB (pCREB), synaptophysin—and a biomarker with high specificity for post-TBI damage, ubiquitin C-terminal hydrolase-L1 (UCH-L1). Mice given immediate saline treatment exhibited higher densities of pCREB in the brain than those given plasma (p = .044); pCREB also trended higher in sham mice than those with rmTBI (p = .084). No significant differences were found when assaying for total CREB, synaptophysin, or UCH-L1. An enzyme-linked immunosorbent assay (ELISA) was also conducted to quantify peripheral changes in the circulating blood plasma of rmTBI and sham mice that were administered the delayed plasma or saline treatment. Although growth differentiation factor 11 (GDF11) has been previously recognized as a prime circulatory factor in delivering the benefits of plasma treatment, the results demonstrated no differences in the GDF11 concentration between plasma-treated and saline-treated mice, nor between rmTBI and sham mice. A second ELISA assessed the plasma concentration of C-C Motif Chemokine Ligand 11 (CCL11), a potential biomarker for CTE-type pathology. However, there were no observed differences in CCL11 levels between rmTBI and sham mice, nor between mice given delayed plasma and saline treatment at 1 month post-TBI. Further plasma analyses for immediate treatment effects will be conducted in future studies within the Flinn lab. The results of this study suggest that rmTBI incurred at a young age may not produce damaging effects to the brain’s plasticity-related proteins and that biological samples must be collected within the acute window after injury to adequately measure inflammatory changes. Future research should continue to probe for diagnostic/prognostic rmTBI biomarkers that can be observed even at a young age and in late-stage measurements, as mild brain injuries still produce long-term consequences in young athletes
Work It: An Ethnography of Drag Performance in Washington, DC
Full text linkDrag performers are seizing more paying opportunities and making it work “full-time,” reflecting what journalists and scholars proclaimed a "drag boom." Investigating Washington, DC's drag scene, I blend interviews, ethnographic and autoethnographic narratives, and an eclectic range of cultural texts to explore the political-economic dimensions of drag performance. I argue drag performance is post-Fordist work and diagnose how the gig economy, rainbow capitalism and changes in social media have contributed to recent transformations in the scene and prompted new risks and rewards. The work of drag performance can be exploited by business owners, show producers, parents at Drag Story Hour, and politicians to signal virtues of diversity and inclusion. Neoliberal frameworks of identity politics equate visibility to liberation and in effect popularize determinist and racialist strategies for diversity and social justice. Finally, throughout this dissertation I present real-world solutions for activists and drag performers: that gigs should always have formal contracts; representation is not politics; and the freedom of speech will not protect drag performance from being considered adult entertainment
Investigating Extracellualr-matrix Derived Hydrogels from the Lungs of Patients with Idiopathic Pulmonary Fibrosis as an In Vitro Disease Model
Full text linkIdiopathic pulmonary fibrosis (IPF), one of the most common forms of interstitial lung disease, is a poorly understood, chronic, and often fatal condition with only two FDA-approved medications. IPF is characterized by excessive matrix production that leads to the loss of lung architecture and physiology, eventually resulting in death. The primary disease-causing cell in IPF is the fibroblast, responsible for depositing matrix components in the lung interstitium. Understanding the pathobiology of the fibroblast is critical to evaluating and discovering novel therapeutics. Unfortunately, our ability to interrogate this biology in vitro is greatly limited by the effects of tissue culture (TC) plastic on the fibroblast phenotype. Previously, we observed a dramatic change in gene expression when culturing fibroblasts on TC plastic. Additionally, it has been well documented that mechanobiological signals derived from stiff substrates such as TC plastic result in the transition of fibroblasts to an α-smooth-muscle-actin expressing phenotype called the myofibroblast (highly migratory and contractile), thus a masking expression of the other subpopulations of fibroblasts intrinsic and relevant to IPF. To overcome the changes in the mechanobiology from TC plastic and to better understand the fibroblast in the IPF lung microenvironment in vitro, we decellularized lung matrices derived from IPF patients to generate three-dimensional (3D) hydrogels as in vitro models of lung physiology. In this project, we characterize the phenotype of fibroblasts seeded into the hydrogels. First, we validate that our methodology preserves native collagen, and soluble ECM becomes a gel-like material under physiological temperature. We also demonstrate that the mechanical properties of ECM hydrogels are comparable to established hydrogel models. Second, we show that when cultured in ECM hydrogels, IPF fibroblasts display differential contractility compared to their normal counterparts, lose the myofibroblast phenotype, and increase expression of proinflammatory cytokines compared to fibroblasts seeded two-dimensionally (2D) on TC dishes. We validate this pro-inflammatory state in fibroblast-conditioned media studies with monocytes and monocyte-derived macrophages. Finally, we utilize FDA-approved drugs in an ECM hydrogel system to verify and confirm drug testing capability. Our results demonstrate that a Janus Kinase (JAK) inhibitor reduces this system's proinflammatory state. Additionally, we observed that fibroblasts cultured in the IPF ECM hydrogel were responsive to pro-fibrotic and anti-fibrotic compounds. These findings 1) reaffirm the crucial limitations of the traditional 2D plastic TC model and their significant effect on fibroblasts, 2) add to a growing understanding of the lung microenvironment effect on fibroblast phenotypes, and 3) shed light on the therapeutic potential of intervention in fibroblast-immune cell crosstalk in the IPF lung
THE POLICE STATE LEGACY, SECURITY SECTOR GOVERNANCE AND REFORM (SSG&R) WITHIN A P/CVE* FRAMING IN TUNISIA: LOCAL OWNERSHIP AT A CRITICAL JUNCTURE
Full text linkThirteen years after the fall of the Ben Ali regime and the advancement of the “post-police state” narrative in Tunisia, international actors invested in Security Sector Governance and Reform programs (SSG&R) with a Preventing and Countering Violent Extremism (P/CVE) framing. However, the modalities of local ownership within this set of programming are complex, given the programs' security dimension, the police state legacy, and pre-existing socio-political configurations. Through multifaceted and multi-method research, this study contends that SSG&R tends to be driven primarily by the excessive preoccupation with the security of Western ideals with little consideration of local agency, power struggles, socio-political structures, and historical legacies. Suggesting a universalism and the existence of templates in SSG&R for P/CVE. While promoted as a positive case, the SSG&R process in Tunisia remains stalled, focused on the professionalization of the police forces while overlooking citizen participation and oversight, and mechanisms of accountability. While the literature on the local turn continues to argue for more meaningful engagement with local stakeholders, their knowledge, and priorities, it remains critical mainly of how the localization agenda is implemented. Building on comprehensive research and analysis, this study concludes that local ownership in itself is a securitizing discourse that patronizes the ‘local’ by imposing Western-centric conceptualizations of it. It undermines the complexity and agency of locals and views human security as linear and secondary to hard security
AI-assisted Software Analysis and Code Repair
Full text linkProgram analysis and code repair are fundamental to uncovering and patching vulnerabilities. However, achieving precision remains challenging when analyzing large-size and intricate software. An inevitable trade-off exists between software complexity and the scalability of analytical tools, causing traditional program analysis to suffer from efficiency or false alarm issues. With artificial intelligence (AI) becoming increasingly popular, applying AI techniques to tackle such complex issues has attracted significant attention. Consequently, a key research question has emerged: how can AI’s exceptional understanding capabilities in natural language and images be transferred to programming languages? In this dissertation, I explore three paradigms that combine AI and program analysis techniques to boost performance in distinct analysis and patching tasks. First, we present BinProv, a compilation provenance identification tool that reveals the impact of compilation on binary code. Compilation provenance refers to the compiler and optimization levels, which hinder analysts from intuitively comprehending the encoding semantics in binary code. Traditional methods rely on reverse engineering (e.g., disassembling or decompilation) to restore code semantics; however, these methods are often inaccurate. To address this, BinProv leverages embedding models to learn the program encoding semantics and avoid the inaccurate disassembling process. Second, we introduce BinGo, a graph-learning-based security patch identification tool that distinguishes security patches in binary code. The security patch is critical to fixing vulnerabilities, but it is hidden in massive, silently released code changes. Traditional program analysis usually involves extracting control flow and data flow graphs to abstract the program’s structure. We use the graph model to delve deeper into the program’s structural semantics. The graph model can understand code functionality by learning from multiple graphs at scale. Third, we propose PathFix, an automated program repair solution that combines the static-analysis-based framework and the large language model (LLM) to understand the comprehensive program semantics and repair the buggy program. Automated Program Repair (APR) is a compound task requiring multiple steps, such as precise correctness specification and patch synthesis, which are challenging to achieve using only traditional program analysis or LLMs separately. Traditional methods have static-analysis-based repair workflow but limited scalability. Meanwhile, LLM can offer promising performance on code behavior summarization and code generation tasks. Therefore, we employ a static-analysisbased framework to decompose steps and guide LLM in finding feasible repair paths and generating appropriate patches. Through extensive experiments, we observe that these AI-assisted solutions effectively improve performance in the three tasks. They also pave the way for further exploration of combining AI and program analysis techniques to enhance software security
Student–Teacher Relationships from Kindergarten to 3rd Grade for Latine Students in Dual Language Programs
Full text linkStudent–teacher relationships in early education have been investigated widely. The current thesis expands on this literature by exploring these relationships in the unique context of Dual Language Education (DLE) K-3 programs and by examining Latine students. I explore student–teacher relationships as a function of student–teacher ethnicity-matching in Spanish-English, two-way immersion programs across grades K-3, where some students have one teacher and others have two. Participants included 33 teachers and 203 students in kindergarten through 3rd grade in Spanish-English two-way DLE immersion classrooms in North Carolina. Relationships were measured via teacher report with the Student–Teacher Relationship Scale (STRS, Pianta, 2001a). The main research questions were: 1) What is the quality of student–teacher relationships for all students, and does this vary by grade (in grades K-3)?, 2) Do student–teacher relationships vary as a function of student ethnicity and teacher ethnicity, and if so, does this vary across different grades?, 3) Do student–teacher relationships vary as a function of ethnic match between student and teacher?, and 4) Specifically for Hispanic students, how much does having a Hispanic teacher influence their student–teacher relationship quality? My hypotheses were that students and teachers that share the same ethnicity, and those in earlier grades, will have higher closeness scores (and lower conflict) than those who do not share the same ethnicity or are in later grades. Additionally, I expected Hispanic students to have closer relationships with their teachers than White students.
Results indicate that student–teacher closeness for all students was highest in kindergarten and lower for subsequent grades, with Black students showing particularly high closeness in kindergarten but steeper declines by third grade compared to other students. Boys had more conflictual relationships with their teachers than girls, and Hispanic/Latine English teachers had more closeness with their students than teachers of other ethnicities. Students with two teachers had lower closeness with their teachers than students with only one teacher. Importantly, ethnic match between student and teacher was not associated with teacher closeness or conflict in the current study’s TWI settings. Implications of these findings include ethnic match not being as important as the number of teachers for student–teacher relationship quality, and that, overall, TWI programs are efficient at producing an inclusive environment for families of diverse ethnicities
The Emergence and Consolidation of Conflict Economies: The Case of Syria (2011-2017)
Full text linkThis dissertation examines the multifaceted emergence of conflict economies within the intricate socio-political landscape of Syria. Focusing on key independent variables – authoritarian governance dynamics, widespread violence, frail national economy, extensive external intervention, and the fragmentation of territorial sovereignty—I argue how these factors intertwine to shape and consolidate the conflict economies in Syria between 2011-2017. This dissertation offers a nuanced understanding of the emergence and consolidation of conflict economies in Syria as well as set the stage for more in-depth research regarding each of the variables addressed therein. By highlighting the interconnected dynamics at play, it contributes to broader discussions on conflict economies, state fragility, and the complexities of economic resilience in intervention-heavy conflict-affected regions
Investigating Hemodynamic Mechanism Contributing to Cognitive and Mobility Impairment in Asymptomatic Carotid Artery Stenosis
Full text linkAsymptomatic Carotid Artery Stenosis (ACAS) is a medical condition where the major blood vessel supplying oxygen rich blood to brain becomes narrowed. Although ACAS patients do not have a history of stroke or transient ischemic attack, these patients often suffer from cognitive and mobility dysfunction. Carotid research has historically focused on the frequency and risk factors for stroke. Currently, we do not know what interventions are needed to mitigate the impact of cognitive and mobility dysfunction. To address this, we must investigate the mechanisms behind these impairments as they pose a significant public health concern. The carotid stenosis impacts the blood supply to the brain. Circle of Willis plays a vital role in providing collateral blood flow to the arteries which supply blood to the two cerebral hemispheres. The objective of the research is to investigate the impact of anatomical variations in the Circle of Willis on the cognitive and mobility dysfunction in ACAS population. The next part of dissertation describes the impact of collaterals from the ophthalmic artery on the carotid Doppler velocities to categorize a hemodynamically significant stenosis. The presence of collateral from ophthalmic artery could be markers for high pressure drop across a carotid stenosis. This could help select a subgroup of patients at high risk of carotid atheroma rupture due to pressure drop at stenosis. Currently the most popular and profitable modalities to study the hemodynamic mechanism are limited to Magnetic Resonance Imaging, Single Photon Emission Computed Tomography and Positron Emission Tomography each of which have their own limitations. Significant limitations are the lack of portability and the inability to capture hemodynamic responses during mobility functions. Therefore, we need a method to assess real-time changes in cerebral oxygenation during cognitive and mobility tasks in ACAS patients. To address this, I utilized functional Near-Infrared Spectroscopy (fNIRS) to monitor real-time hemodynamic changes during these tasks. Finally, I am investigating the technical challenges associated with measuring cerebral oxygenation during the fNIRS data collection
Ground-based Light Curve Follow-up Validation observations of TESS object of interest TOI 5944.01
Full text link“The Transiting Exoplanet Survey Satellite (TESS) mission, launched in 2018, aims to discover exoplanets by observing transits across bright stars. Using the transit method, TESS collects planetary properties such as radius, mass, and orbital period which are all used to identify potentially habitable planets. In this study, we present ground-based follow-up observations of TOI 5944.01 using data collected at George Mason University Observatory. AstroImageJ was used to do tasks such as data reduction and light curve generation and a NEB check was conducted to indicate a potential false-positive that did not deny the possibility of a false-positive. The analysis calculated a transit depth of 15.53 parts per thousand, closely resembling TESS’s reported 15.16 parts per thousand. TOI 5944.01 was classified as a Hot Jupiter due to a radius 12 times of Earth, an orbital period of 5.94 days, and a temperature of 947 K. However, further investigation is required to confirm a true transit due to the possibility of a Nearby Eclipsing Binary and assess its habitability.