113588 research outputs found
Sort by
‘Learning To Talk to Generative AI Chatbots’: A Corpus Study of Generative AI Prompts, an Emerging Genre for AI Literacy
No full textIn this dissertation, I respond to existential questions about what the role of human writers will be in a world where generative AI (or GAI) tools like ChatGPT produce texts with limited human intervention. In the face of such fears about potentially constrained human rhetorical agency, one of the ways in which writers have been negotiating their agencies with GAI tools is through a new type of writing called “GAI prompts,” which refer to instructions that writers compose to elicit desired outputs from ChatGPT-like tools. Many early adopters of GAI in writing studies and allied fields have been publishing innovative rhetorical experiments in GAI prompt writing. The ability to compose prompts has also been advocated as a key skill within emerging frameworks for ‘AI literacy’ that seek to guide the adoption of GAI tools in an ethical and human-centered manner. However, due to the nascent nature of this emerging form of writing, it remains undertheorized. To address this, I draw on genre theory, corpus methods, and virtue ethics to compile and study a corpus of GAI prompts published by early adopters of GAI in highly visible domains like journals and books in writing studies and allied fields. By analyzing this corpus using a mixture of qualitative and quantitative techniques, I provide a data-driven description of emerging genre characteristics of GAI prompts and present implications for how those data-driven descriptions can be used to support pedagogy, research, and UX design work in rhetoric, composition, and technical communication.Originally embargoed until 04/25/2030; released 11/19/2025 per author's request
Comparing Rhamnolipid-Metal Interactions: Effect of Rhamnolipid Structure Modifications
No full textGlycolipids are bio-inspired surfactants with promising potential for metal remediation and reuse technologies. They offer an environmentally friendly, low-cost alternative to current techniques that provides a solution to growing industrial demand for metals. Metal recovery from contaminated environments, such as mining impacted waters, is critical due to their inherent economic value, and ecological and human health impacts associated with metal toxicity. This study focuses on rhamnolipids, a type of glycolipid originally produced by Pseudomonas aeruginosa. These rhamnolipids can now be chemically synthesized in large-scale batches that can be structurally modified to explore a range of metal-binding affinities. The binding affinities of biologically-produced rhamnolipid with a wide variety of metals has been investigated previously, and they are expressed as log β values. In the first part of this study, the binding affinity of synthetic Rha-C10-C10 rhamnolipid was investigated with silver (Ag+) and scandium (Sc3+), with results showing relatively low complexation for these metals (log β = 2.36 and 2.60). A second part of the study was to compare the binding affinities of Rha-C10-C10 with two newly designed rhamnolipids; 2C-Rha-C14, which has a modification to the sugar-lipid linkage that adjusts the size of the metal-binding pocket, and Rha-Phos-C13, which modifies the functional group, allowing for low-pH binding capabilities. Two metals were used in this comparison, lead (Pb2+) which has been previously studied, and gallium (Ga3+) which has not. Complexation of Pb2+ with Rha-C10-C10 was strong (log β = 10.01) as previously observed. Complexation of the 2C-Rha-C14 was similar to the Rha-C10-C10 (log β = 10.085). However, complexation of Pb2+ by the Rha-Phos-C13, measured at neutral pH was much lower (log β = 6.07). For Ga3+, binding by Rha-C10-C10 was low (log β = 0.57), while binding with 2C-Rha-C14 was moderate (log β = 4.92). However, under acidic conditions, the Rha-Phos-C13 showed increased binding of Ga3+ (log β = 5.41). These results with Rha-Phos-C13 represent the first report of successful metal binding with a rhamnolipid at low pH. The third part of this study characterized the binding of aluminum (Al3+) with Rha-Phos-C13 which revealed that there were strong interactions between the buffer used in the system and Al3+. This led to an investigation of the impact of changing the buffer concentration and the addition of methanol as a solvent. Results revealed higher binding with the standard buffer concentration (log β = 3.10) in comparison to the lower buffer concentration (log β = 1.71). Overall, this study contributed toward a deeper understanding of glycolipid-metal interactions under a variety of experimental conditions. These findings have implications for the continued development of glycolipid-based systems for selective metal recovery, with strong potential for both environmental and economic benefit.Release after 05/21/202
Compound Events and Exposure Science: Characterizing Mixtures and Assessing Cumulative Risks Using Participatory Research Methods
No full textEnvironmental justice (EJ) communities face disproportionate contaminant exposure due to systemic inequities, industrial proximity, and inadequate regulatory protections, with climate change further exacerbating these risks. This work explored environmental exposure in a diverse range of environmental justice communities, including border, mining and urban communities. This dissertation employed a transdisciplinary, multi-contaminant, and multi-media approach to assess environmental exposures, integrating environmental monitoring, chemical mixture analysis, climate-driven exposure pathways, and community-based research. Rather than focusing on a single pollutant, it examined metal(loid)s, per- and polyfluoroalkyl substances (PFAS), polycyclic aromatic hydrocarbons (PAHs), and dioxins across soil, rooftop-harvested rainwater, settled dust, and drinking water, providing a comprehensive understanding of contaminant distribution, transport, and human exposure pathways. Therefore, this work utilized a systems thinking approach to understand the interconnectedness between contaminants, communities, climate change and the environment. The first study applied a community-engaged research approach to assess metal(loid) contamination in drinking water sources along the Arizona-Mexico border, a region experiencing severe climate change impacts, including prolonged droughts, extreme heat, reduced groundwater recharge, and intensified monsoon flooding. These challenges are further compounded by the complexities of binational environmental management, affecting water quality, availability, and long-term sustainability. Therefore, residents rely on pipas (water trucks), private wells, and public water systems. This study used the Social Determinants of Health (SDOH) framework to examine the relationships between measured contaminant levels, water source types, and public perceptions of water quality. Although most drinking water samples met regulatory limits in this study, arsenic and lead concentrations exceeded the maximum contaminants level goals. Additionally, the study finds that perceptions of water quality often do not align with measured contamination levels, indicating a need for improved risk communication strategies and increased public health interventions. The second study employed a co-created community science approach to examine PFAS contamination in a mining-impacted EJ community affected by wildfires and post-fire flooding. This study evaluates potential PFAS release from mining operations, while assessing the role of extreme climate events in contaminant redistribution. Soil samples from flood-affected residential areas had significantly elevated PFAS concentrations, suggesting that wildfire and flood events facilitate contaminant transport. These findings highlight the importance of long-term monitoring and the need for regulatory strategies that account for climate-driven remobilization of persistent pollutants. The third study continued the work from the previous co-created community science study and applies cumulative risk assessment methodologies to evaluate co-exposure to PAHs, dioxins, and PFAS in post-wildfire and flood-affected soils and indoor dust. Using Monte Carlo-based probabilistic risk modeling, this study quantifies both carcinogenic and non-carcinogenic risks associated with chemical mixtures in an EJ community. The results show that PAHs contribute the most to estimated cancer risks, largely due to their release from wildfires, which generate combustion byproducts that accumulate in soils and dust, and flooding, which remobilizes PAHs from contaminated industrial sites, roadways, and degraded infrastructure. Meanwhile, PFAS contributed the most to the non-carcinogenic hazard indices. This study highlights the limitations of single-chemical risk assessments and demonstrates the necessity of mixture-based exposure models to better characterize environmental health risks. Given the complex interactions between contaminants, the use of Toxic Equivalency Factors (TEFs) is essential for evaluating the combined toxicity of as different compounds within the same chemical class exhibit varying degrees of toxicity. The fourth study used co-created community science data to investigates metal(loid)s contamination in rooftop-harvested rainwater (RHRW) in four Arizona EJ communities using the Pollution Load Index (PLI) to quantify cumulative contamination. Results indicate that mining-impacted communities had the highest contamination levels, with monsoon season contributing to increased metal(loid) deposition through atmospheric transport and surface runoff. The findings emphasize the need for enhanced regulatory oversight and public health guidance to ensure the safe use of harvested rainwater in water-insecure regions. This dissertation advances environmental exposure science by integrating cumulative risk assessment, climate-responsive exposure models, and community-based participatory research. By characterizing the distribution and health risks of complex contaminant mixtures in EJ communities, this work provides critical data for improving environmental health assessments and policy decisions. The findings support the development of adaptive regulatory frameworks that incorporate climate change considerations and emphasize the importance of community-engaged research in shaping public health interventions. Future research should build on these findings to refine exposure models, improve mixture-based risk assessment methodologies, and develop targeted mitigation strategies that reduce environmental health disparities in vulnerable populations. Release after 11/16/202
Development of a Multi-Slit Based In-Vivo Confocal Ophthalmoscope
No full textIn clinical practice, diagnosis of corneal ulcers (keratitis) requires invasive corneal scrapes and long waits for tissue pathogen culture development. This current standard of care leads to a negative patient experience and requires highly trained clinicians to conduct examinations. In vivo confocal microscopy (IVCM) has been shown to be promising for accurately diagnosing patients with microbial infection of the cornea. With the goal of lowering the cost and increasing the accessibility of IVCM, we created a new IVCM with an LED source, CMOS sensor, and water-immersion objective lens. The IVCM prototype uses illumination and detection slits, each of which has multiple periodic openings to increase the signal level while retaining high resolution. The device was created at a fraction of the cost with comparable resolution (2.5 µm lateral and 4.6 µm axial resolution) and significantly faster imaging speed (100 frames per second) to the current commercially available IVCM.Release after 05/23/202
Enzyme Filamentation: A Focus on Mammalian Glutamate Dehydrogenase
No full textEnzyme filamentation is an emerging regulatory mechanism influencing enzymatic activity, stability, and metabolic organization. Chapter one explores the structural and functional significance of enzyme filaments and how higher-order assemblies modulate reaction kinetics and metabolic flux. This chapter provides an overview of known filament-forming enzymes and their physiological roles, including but not limited to, acetyl-CoA carboxylase (ACC), CTP synthase (CTPS) and inosine-5-monophosphate dehydrogenase (IMPDH).Chapter two focuses on glutamate dehydrogenase (GDH), a key enzyme in nitrogen metabolism that catalyzes the reversible conversion of glutamate to α-ketoglutarate and ammonia. GDH serves as a critical metabolic node, linking amino acid catabolism to the tricarboxylic acid (TCA) cycle and responding to cellular energy demands. This chapter examines what is known about the structure, regulation, and function of mammalian GDH in hexameric and filamentous forms. Chapter three is centered on the kinetics of mammalian GDH filamentation and its biochemical and functional implications (with raw data presented in Appendix A). This chapter details the effect of GDH filaments on enzymatic activity, and their regulatory significance. Chapter four reports on two new cryo-EM structures of filamentous bovine GDH and structure-function analysis combining the kinetics data from Chapter three. This study provides new knowledge to the growing field of filament-forming enzymes and enhances our understanding of the biochemical and enzymatic nature of mammalian GDH1 filamentation. Finally, chapter five outlines future directions, proposing the next steps to deepen our understanding of GDH filamentation and its potential for therapeutic applications. These findings provide a foundation for future structural studies that would uncover the functional significance in vitro and in vivo.Release after 05/26/202
Resilient and Risk-Averse Network Systems
No full textGraphs are versatile modeling tools capable of effectively representing real-world systems by capturing individual components and their intricate interactions. The main objective of this research is to formulate and develop efficient solution methodologies for graph-theoretical problems involving topologically stochastic information appearing in various forms. Weutilize a stochastic programming framework grounded in coherent risk measures to identify minimum-risk graph structures under stochastic vertex and/or edge weights. First, we consider a network flow problem and propose an approach for developing resilience metrics, where resilience refers to the network’s capability to restore optimal or near-optimal operations following unforeseen (stochastic) disruptions in topology or operational parameters. We illustrate this approach through two examples: the resilient maximum network flow problem and the resilient minimum cost network flow problem. Specifically, these network flow problems aim to achieve resilience against unpredictable losses of network arc capacities by preallocating resources to restore, at least partially, the capacities of arcs. Additionally, we demonstrate that the proposed formulations of resilient network flow problems can be interpreted as ``network risk measures'', possessing properties analogous to convex risk measures. Efficient decomposition algorithms are developed for solving both the resilient maximum network flow problem and the resilient minimum cost network flow problem. Furthermore, we analyze network flow resilience in relation to network structure by conducting studies on three distinct types of network topology: uniform random graphs, scale-free graphs, and grid graphs. Next, we address the regulator's perspective on risk, which concerns regulators or large asset holders who seek to manage the risk of their assets through some investments. A central feature of this perspective is that losses can be “suppressed” by allocating additional resources, meaning that committing zero resources leaves losses unchanged. To illustrate this, we examine two well-known graph problems: the minimum vertex cover and the spanning tree problem. We present mathematical programming formulations for both and employ a stochastic programming framework grounded in coherent risk measures to identify minimum risk graph structures under stochastic vertex or edge weights. After establishing that the decision versions of these problems are NP-hard, we propose a combinatorial branch-and-bound algorithm and compare its performance with an equivalent mathematical programming approach on randomly generated networks. Lastly, we examine graph strength as a robust measure of connectivity. Graph strength quantifies a network's resilience by determining the number of edge-disjoint spanning trees that can be embedded within it. By leveraging this concept, we gain insights into designing networks capable of maintaining functionality despite edge failures, ensuring multiple disjoint paths between any two vertices. We propose a cutting-plane mathematical programming method for computing graph strength, where constraints are iteratively generated by solving a minimum spanning tree problem. Additionally, we investigate the problem of identifying a subgraph with at least a prescribed strength and the largest possible cardinality. We show that while deleting edges from a graph cannot increase its strength, removing vertices can potentially enhance the strength of the remaining subgraph, thereby making the problem well-defined. We prove that finding the maximum subgraph of a given strength is solvable in polynomial time and present the corresponding algorithm. Finally, we provide numerical experiments conducted on a diverse set of real-world graphs to demonstrate the computational performance of the proposed algorithms.Release after 05/16/202
Soil Microbial Adaptations to Climate Disturbance: Integrated Multi-Omics Insights from Permafrost and Arid Ecosystems
No full textClimate change is intensifying temperature and precipitation variability, challenging microbial communities and ecosystem functioning. Interactions between microbiomes and metabolites are central to biogeochemical cycling, yet how these interactions drive greenhouse gas emissions during ecosystem transitions remains poorly understood. To address this, we applied an integrated multi-omics approach—combining metagenomics, metatranscriptomics, and metabolomics—across two climate-sensitive ecosystems: thawing permafrost peatlands and monsoon-influenced arid soils.In Stordalen Mire, Sweden, we analyzed microbial and metabolite composition along a permafrost thaw gradient using genome-resolved metagenomics and high-resolution Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS), guided by community assembly theory. We found divergent assembly processes between microbial taxa and metabolites in response to the same environmental drivers, challenging assumptions in trait-based microbial models. Feature-level analysis revealed associations between microbial taxa, metabolite profiles, and variations in porewater CO₂ and CH₄, highlighting the importance of microbe–metabolite interactions in greenhouse gas flux. We investigated arid soil microbial communities, which endure extreme and rapidly fluctuating environmental conditions, particularly during monsoon transitions. Contrary to the assumption of widespread dormancy, our multi-omics analyses show that microbial populations remain metabolically active, exhibiting dynamic functional redundancy. Communities rapidly reconfigure gene expression and metabolite profiles in response to moisture and nutrient pulses. This functional plasticity enables continuous ecosystem processes and reveals a resilience strategy previously underappreciated in arid soils. Furthermore, we observed consistent partitioning of metabolic functions across ecological preference groups, supporting ecosystem functioning despite environmental extremes. Altogether, our findings demonstrate that microbial responses to climate perturbations are shaped by distinct, ecosystem-specific processes linking microbial function, metabolite dynamics, and environmental change. By integrating high-resolution molecular tools with ecological theory, this work provides new insight into the metabolic mechanisms driving ecosystem resilience and greenhouse gas emissions, offering a predictive framework for understanding microbial contributions to climate feedbacks in a warming world.Release after 05/09/202
Statistical Methods for Temporal Biomedical Data
No full textTemporal biomedical data provide critical insights for disease identification, prevention, and treatment by capturing how biological and clinical variables evolve over time. Understanding temporal patterns in patient health records, biomarker fluctuations, and disease progression enables early detection of emerging conditions, informs timely medical interventions, and supports personalized treatment strategies. However, existing statistical methods face challenges in effectively analyzing these complex datasets. This dissertation addresses three key challenges in temporal biomedical data analysis: (1) identifying dynamic, time-dependent effects of risk factors on the outcome while avoiding aggregation that can potentially obscure biologically relevant associations, (2) modeling temporal trends in the presence of excessive zeros, particularly in sparse datasets like single-cell RNA sequencing (scRNA-seq), and (3) detecting and summarizing clinically meaningful temporal patterns.
The first project introduces TIDE, a novel method for individual-level scRNA-seq analysis that enhances the accuracy and resolution of dynamic differentially expressed (DE) regions along cell pseudotime. Unlike traditional approaches that focus on global DE gene identification, TIDE provides a more granular view of gene expression changes within regions of pseudotime, making it particularly valuable for studying cellular differentiation, disease progression, and treatment responses. The method employs gene-specific functional principal component regression (FPCR) to model gene expression in relation to the outcome, followed by a permutation test on the estimated coefficient function to obtain time-specific p-values. TIDE demonstrates well-controlled type I error and decent statistical power across different sample sizes and group imbalances. Additionally, TIDE’s flexibility allows for branch-specific analysis within pseudotime structures and the incorporation of individual-level covariates, making it adaptable to diverse experimental designs. While our implementation uses TSCAN-derived pseudotime, TIDE can integrate trajectories from other methods, further enhancing its applicability across single-cell studies. The second project presents a novel statistical method ZISS to handle zero inflation and over-dispersion in scRNA-seq datasets. ZISS consists of two parts: a time-dependent function that models the probability of excessive zeros and another that estimates the mean of the zero-inflated Poisson distribution. To address the challenge of irregularly observed temporal data in existing methods, ZISS employs smoothing spline ANOVA for mean function estimation. B-spline basis functions are used to adaptively model the dynamic zero probabilities over time. ZISS outperforms traditional models in challenging conditions, achieving the lowest mean square error of 0.169 and the smallest standard deviation of 0.009. It excels in high zero-inflation data analysis, and adaptively differentiates between true zeros and technical zeros in scRNA-seq data in simulation studies. Additionally, ZISS maintains consistent accuracy across varying levels of over-dispersion and is versatile enough for applications in other domains involving temporal or spatial data with excessive zeros. The third project applies a cluster-based method to analyze temporal medical data, i.e., corrected QT (QTc) variability, to evaluate disease severity subtypes that complement the severity defined by the traditional metric in patients with obstructive sleep apnea (OSA). We identify a high-risk OSA subtype that more precisely distinguishes cardiovascular disease (CVD)-related mortality risk. Unlike previous metric based on apnea-hypopnea index (AHI) that failed to differentiate mortality risks between moderate and severe OSA, our clustering approach reveals a distinct severe subtype (pattern 2) associated with significantly higher CVD-related mortality compared to other OSA subtypes, including those with mild OSA, no OSA, and those categorized under cluster pattern 1. Furthermore, the severity subtypes detected from temporal records of QTc variability moderate the relationship between the mean QTc and CVD-related mortality, after adjusting for age and gender. This framework provides a more comprehensive understanding of QT variability across different OSA severity levels, offering a novel framework to explore the relationship between QT dynamics and OSA severity.Release after 11/21/202
High-Performing Upland Cotton Shifts Root-Associated Microbiomes Under Water Limitation
No full textWater scarcity significantly threatens cotton productivity, a challenge amplified by climate change and increasing competition for limited water resources. As a major source of natural fiber, cotton's resilience to drought stress is essential for maintaining productivity and supporting global textile production. However, the mechanisms underlying this resilience, particularly the responses of root-associated microbial communities that may influence plant drought stress responses, remain unclear. Here, we quantified the plasticity of microbial communities associated with roots of six cotton cultivars grown under water-limiting and well-watered conditions in a hot, arid environment. The highest-yielding cotton cultivars markedly shifted their root microbial communities between irrigation treatments, whereas low-yielding cultivars were less responsive. Microbiome shifts in high-performing varieties suggest that these plants may leverage symbiotic relationships to cope with water limitation. This study links microbial communities and the performance of cotton and highlights the potential for leveraging these relationships to improve crop resilience in water-limited environments
Delivered: Living and Working with Algorithms
No full textLabor platforms—digital platforms that algorithmically mediate between companies, workers, and consumers—have caught the attention of organizational scholars, sociologists of work, and science and technology scholars. Attention has been given primarily to algorithmic control—that is, the usage of algorithmic systems to manage workers at a distance. Less attention, however, has been given to how the algorithmic systems embedded in labor platforms are experienced on the ground by workers who are embedded in specific cultures and economies. Without scholars paying attention to how algorithmic systems in labor platforms unfold in situ, these systems remain conceptually static and closed. In this dissertation, I address this gap by moving from algorithmic control to algorithmic practice. In particular, it investigates the practices, interpretations, and contestations that workers have with regard to algorithmic systems. To do so, I draw from 15 months of ethnographic fieldwork spent working alongside platform-based food delivery couriers in Mexico City, 75 semi-structured interviews with platform-based food delivery couriers, and online ethnography conducted in online Facebook communities. This dissertation is organized in three articles. The first article draws primarily from interviews to address how couriers evaluate platform-based delivery gig work in relation to other available options in the urban labor market in Mexico City. Evaluation and perception of platform-based work is dependent upon workers’ past employment trajectories. This article moves away from a conceptualization of homogenous reception to address workers’ heterogeneity. The second article draws from participant observation as a food delivery courier, ethnographic observations at bases, and online ethnography in Facebook forums to address how the labor process in food delivery platforms is sedimented across spaces exhibiting different levels of virtuality. Rather than concentrating solely on algorithmic control, this article takes seriously the agency of couriers and details how they engage in practices of consent and resistance as they interact with algorithmic technologies embedded in food delivery platforms. The third article also draws from participant observation as a food delivery courier, ethnographic observations at bases, and online ethnography in Facebook forum in order to address the gap that exists between technological design of food delivery platforms and local economic practices, particularly the saliency of cash in Mexico City. Given that cash administration requires physicality, this task is delegated from platforms to couriers. This results in an added layer of algorithmic control given that platforms must ensure that couriers are completing this task adequately. Moreover, this task is not experienced uniformly among workers as existing socioeconomic divides shape this delegated task. This article addresses how algorithmic control does not follow a universal template and how local economic contexts shape technologies themselves. Across these three articles, this dissertation offers insightful contributions to our understanding of the intersection between technological diffusion and digitally mediated labor control. Despite the world of work experiencing an era of technological convergence, local contexts shape how these digital technologies are interpreted, navigated, and contested by workers on the ground