application/pdf
The design of DNA-containing biomaterials for forensic science
Access to human blood for forensic research and training in bloodstain pattern analysis (BPA) can be difficult due to many ethical, safety and cost concerns. Mammalian blood alternatives can be sourced, especially from local and willing abattoirs, but some concerns remain, and the added difficulties of high variation and species-specific differences in cellular components pose other issues. Therefore, synthetic alternatives to human blood provide practical options for the BPA community. This thesis explores the use of alginate hydrogels as a base material for forensic blood substitute (FBS) development. Hydrogels are first explored as a suitable environment for DNA stability and functionality and compared to other polymer systems. The ability of DNA to remain intact while undergoing electrospray ionization (ESI) is also investigated. The FBS design focuses on mimicking the fluid properties and genetic capabilities of whole human blood – a material not developed in FBSs previously. ESI was used to develop microparticles (MPs) that serve as cellular components of human blood (the red blood cells – RBCs, and white blood cells – WBCs). The microparticles were ionically crosslinked using calcium to provide small MPs (RBCs) or covalently crosslinked with functional DNA to provide larger WBC-like functional particles. The integration of these novel MPs into alginate-based materials is optimized and their use in BPA scenarios is explored. The FBS is tested in BPA scenarios of dripping experiments, impact patterns, and the ability to extract and amplify the contained DNA. In addition, the stability (or shelf-life) of the FBS was also assessed. The FBS exhibited similar spreading ratios to blood and demonstrated feasibility in use for impact angle (a) determination and impact pattern creation. Importantly, the DNA contained within the FBS could be processed with analogous protocols used in DNA evidence processing, enhancing its applicability to BPA research and training.
Author Keywords: Alginate hydrogels, Bloodstain pattern analysis, Electrospray ionization, Forensic blood substitutes, Forensic materials, Synthetic DNA design
Optimized Large Language Model for Hate Speech Detection
Recent developments in Artificial Intelligence (AI), particularly Large Language Models (LLMs), have provided powerful tools for Natural Language Processing (NLP) tasks like sentiment analysis. However, their fine-tuning and deployment present challenges, specifically in terms of computational efficiency and high training costs. To address these challenges, this work applies optimization techniques such as Quantized Low-Rank Adaptation (QLoRA) for parameter-efficient fine-tuning, followed by Generalized Post-Training Quantization (GPTQ) on the Llama 3.1 LLM. To evaluate these optimizations, we apply the model to a practical task: hate speech detection, using a curated dataset comprising of X (formerly Twitter) posts. Overall, the optimized model achieved a 67% reduction in size along with significant improvements in classification accuracy and inference speed compared to the base model.
Author Keywords: Generalized Post-Training Quantization, Large Language Models, Low-Rank Adaptation, Parameter-Efficient Fine-Tuning, Quantized Low-Rank Adaptation
The effect of oncology camp on the psychosocial health of childhood cancer patients
Objectives/purpose: This study examined the psychosocial health (including resilience, hope, social functioning, mental wellbeing, and stress) of childhood cancer patients attending a recreational oncology camp (ROC).Methods: Childhood cancer patients enrolled for ROC participated in a survey on the first (T1) and last (T2) day of camp, and 3 months post-camp (T3). This survey included the: Child and Youth Resilience Measure (CYRM), Children's Hope Scale (CHS), Social Provisions Scale (SPS-5), and Short Warwick-Edinburgh Mental Wellbeing Scale (SWEMWBS). Afternoon saliva samples were collected at T1 and T2 to determine cortisol (ELISA). Results: CYRM, SPS-5, and SWEMWBS scores were high but did not differ between timepoints. CHS scores at T3 were lower compared to T1 and T2 (F=9.388, p=0.008). Salivary cortisol levels were within normal ranges. Conclusion/clinical implications: Childhood cancer patients have high levels of positive psychosocial variables and normal stress levels while attending ROC.
Author Keywords: cancer, children, hope, oncology camp, psychosocial health
Connecting Entanglement Entropy Growth and Local Integrals of Motion in the Disordered Fermi-Hubbard Model
We study a one-dimensional Fermi-Hubbard model with disorder in charge and spin degrees of freedom. We calculate the time dependence of the entanglement entropy. While previous research on disordered interacting systems has typically focused on systems with either charge or spin, our model enables us to explore the interplay between charge and spin in shaping the behavior of entanglement. We use a method that identifies optimally local charge- and spin-specific integrals of motion. We ask how the locality level of these integrals of motion influences the capacity of low-order terms in the l-bit Hamiltonian to capture the entanglement entropy. Our results show that increasing the locality level improves the accuracy of low-order terms in capturing entanglement entropy dynamics. With equally strong charge and spin disorder, the behavior of the entanglement entropy closely resembles that observed in single-degree-of-freedom systems, and the l-bit Hamiltonian truncated at second order accurately captures this behavior.
Author Keywords: Entanglement Entropy, Fermi-Hubbard Model, Hungarian Algorithm, l-bit Hamiltonian, Local Integrals of Motion, Many-Body Localization
Performance of Time Series Interpolation Algorithms in the Presence of Noise
The spectral properties of time series data reveal underlying processes but require complete datasets, often unavailable due to missing values and irregular sampling.This thesis uses a computational simulations framework to evaluate the perfor- mance of the Hybrid Wiener Interpolator [3], a novel method designed to reconstruct nonstationary time series data, thus making said data amenable for spectrum analysis. This research evaluates the Hybrid Wiener Interpolator's ability to handle nonstation- ary data and data gaps, comparing its performance to other interpolation methods under different stationarity and data integrity conditions. The results illuminate the robustness of this interpolator in scenarios typical of scientific datasets, offering a promising approach for enhancing spectrum estimation in the presence of non-ideal data conditions
Author Keywords: ARIMA Models, Data Imputation, Interpolation, Stationarity, Time Series, Time Series Simulations
Simulating Microwave Morphology Dependent Resonances in Aqueous Dimers and Trimers
Microwave resonances in isolated water-based spheres, dimers, and trimers are explored using simulations conducted with COMSOL Multiphysics. The study centers on morphology-dependent resonances (MDRs) and hotspot characteristics in cm-sized objects at microwave frequencies. Monomers subjected to microwave radiation exhibit four distinct resonant modes at specific sizes characterized by electric and magnetic field distributions which correspond to magnetic-dipolar, electric-dipolar, magnetic quadrupolar, and electric quadrupolar resonances, respectively. Dimer configurations reveal intriguing hotspot features, with axial hotspots emerging as a key resonant characteristic. The three fundamental dimer orientations dictate unique resonant behaviors, highlighting the sensitivity of hotspot intensity to orientation changes, but smooth and consistent trends during transitions between them. Investigations into trimer structures, as a more intricate geometry formed by interconnected dimers, reveal the subtle interactions of spheres in a trimer structure. Trimer hotspots largely reflect the sum of isolated dimer hotspot contributions, showcasing the energy conservation with no evidence of a newly formed hotpot. Our results, while arising as a consequence of the particularly high index of refraction of water at GHz frequencies, are generalizable to other length scales (such as nano-photonics), were materials with sufficiently high refractive index and transparency to be found.
Author Keywords: COMSOL simulations, Electromagnetic physics, Microwave frequencies, Morphology-dependent resonance, water-based objects
The ecological, evolutionary, and behavioural context of ultrasonic vocalizations in flying squirrels
Acoustic communication is vital for mediating interactions between individuals and their environment. From echolocating bats to isolated rodent pups, ultrasonic vocalizations are a useful mechanism for producing localized and cryptic acoustic signals. Flying squirrels have been demonstrated to produce almost exclusively ultrasonic calls which is rare in both squirrels and mammals, though the significance of this acoustic range is unknown. To investigate the context of ultrasonic calls in these gliding mammals, I created phylogenetic comparative analyses to compare ecological and morphological traits against call frequencies. I found that nocturnality, a trait entangled with gliding in squirrels, was related to higher-frequency calls. Furthermore, by comparing all gliding mammals and their non-gliding counterparts, I found that gliding mammals produce significantly higher frequencies than these relatives. This form of cryptic communication is likely used to avoid predation, which was further supported by behavioural experiments wherein flying squirrels used significantly higher frequencies when predation risk increased. While high frequency communication was used by most gliders, I found that ultraviolet-induced photoluminescence, another potential form of crypsis, was strongly associated with nocturnality, with half of the tested gliding mammals showing evidence of this unique colouration. While ultrasonic vocalizations are widespread in echolocating bats and echonavigating mammals, I did not find evidence to support echonavigation in flying squirrels. Instead, I found that ultrasonic vocalizations are used in a variety of social contexts and during solitary foraging and exploration. These combined results demonstrate the unique ecological and evolutionary pressures acting on small-bodied, nocturnal gliding taxa and the resulting cryptic behaviours and communication.
Author Keywords: communication, flying squirrels, gliding mammals, Sciuridae, ultrasound, ultraviolet-induced photoluminescence
"Bow Down, Bitches" How Beyoncé's Art Reflects and Contributes to the Notions of Sisterhood, Female Empowerment, and Intersectionality within the Framework of Black Feminist Thought
This thesis examines Beyoncé's art within the context of Black feminist thought, specifically focusing on how her work reflects and contributes to the themes of sisterhood, female empowerment, and intersectionality. A comprehensive analysis of her songs and performances will demonstrate how Beyoncé's art advocates for unity, female empowerment, particularly for Black women, and encourages sisterhood and support. The results reveal that Beyoncé's art serves as a powerful tool to challenge societal norms, address racial and gender inequalities, and advocate for justice, especially in the lives of Black women. Through her music and performances, Beyoncé has become a powerful example of using popular music as a medium for social change and cultural empowerment. This research highlights the significance of her contributions to the ongoing conversations surrounding race, gender, and socioeconomic factors, underlining the powerful influence of her art in encouraging a more inclusive society.
Author Keywords: Beyoncé, Black Feminist Thought, Female Empowerment, Gender inequalities, Intersectionality, Sisterhood
Adsorption of Naphthenic Acids from Oil Sand Process-Affected Water (OSPW) using Synthesized Activated Carbon
This thesis explores the remediation of naphthenic acids (NAs) from oil sands process-affected water (OSPW) using activated carbon (AC) derived from petroleum coke (PC) chemically activated with potassium hydroxide. The research aims to identify the most effective method for the adsorptive removal of NAs by optimizing the use of economically viable KOH quantities and to apply Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (FT-ICR-MS) for species-specific detection and characterization of NAs, crucial for targeting specific NAs in future studies.Prior research focused on single-species adsorption, establishing a foundational understanding of non-competitive adsorption before applying these findings to more complex NA mixtures and OSPW. This study builds upon this foundation, addressing a significant gap in the literature concerning the use of petcoke-derived AC with low KOH ratios and short activation times, which are economically advantageous for large scale applications. In this thesis, a comprehensive investigation into the kinetics and isotherms of NA adsorption on various ACs including PAC (petroleum coke AC), PWAC (pore-widened AC), HAC (heat-treated wood-based AC), and CAC (commercial AC) was conducted. The study specifically examines the adsorption behaviors of seven model NAs, reflecting the diverse molecular structures present in real world OSPW. The research also explores the impact of pore widening techniques on the adsorption efficiency of ACs, hypothesizing that increased mesoporosity enhances the adsorption of NA compounds. The findings demonstrate that FT-ICR-MS is an essential tool for precisely characterizing the NA species in OSPW, revealing that pore-widened ACs significantly improve the adsorption of NAFCs. This thesis contributes to the field of environmental remediation by offering new insights into the optimization of AC for NA removal, emphasizing the importance of surface chemistry and mesoporosity in enhancing adsorption efficiency. The study's outcomes have significant implications for the treatment of OSPW, providing a scalable and cost-effective solution to mitigate the environmental impacts of oil sands production.
Author Keywords: activated carbon, FT-ICR-MS, naphthenic acids, oil sands, petroleum coke, process-affected water
Effect of Polarization on the Band Structure at a Charged Domain Wall in Ferroelectric Materials
The interplay between electron charge, spin, and ferroelectric polarization is under-explored for conducting ferroelectric domain walls (DWs). DWs are interfaces that separate regions (domains) within a material that have different orientations of spontaneous polarization. We investigated the electronic band structure of t2g electrons, confined to 90° charged do main walls (CDWs) in barium titanate (BaTiO3), a prototypical perovskite ferroelectric. A key novel aspect of our study is the explicit inclusion of both orbital and spin degrees of freedom in the Hamiltonian. This leads to an Ising-type spin-orbit coupling (SOC). We constructed a tight-binding (TB) model for t2g electrons that is constrained by symmetries of the DW, including time-reversal, mirror, and rotational symmetries. First-principles density functional theory (DFT) calculations were performed to extract the TB parame ters. Our findings offer new insights into spin-orbit interactions at ferroelectric domain walls and open avenues for their potential use in next-generation electronic and spintronic devices