14277 research outputs found
Sort by
Characterizing iron complexing ligands in aquatic environments using Immobilized Metal Affinity Chromatography
No full textIron plays an active role in aquatic environments, acting as a key micronutrient for micro-organisms such as phytoplankton. Iron is difficult for these organisms to access in oceanic water due to low ambient concentrations. Iron has two redox states, the II redox form is soluble in water under anoxic conditions; however, it is easily oxidized to its III form in the presence of oxygen and precipitates as Fe₂O₃(s). As a result, dissolved iron concentrations are sub-nanomolar in seawater. Microorganisms have evolved strategies to keep iron in solution, namely via the production of siderophores, specialized iron-chelating mole-cules. 99% of the dissolved iron in the ocean is bound to organic ligand complexes, which help to maintain the bioavailability of the metal. These ligands are considered part of dissolved organic matter (DOM), a complex carbon pool containing eclectic water-soluble compounds of varying chemical compositions. The vast majority of DOM is chemically uncharacterized and often referred to as humic substances (HS), which originate from the breakdown of organic matter of biological origin. A subset of HS can bind selectively to iron and retain it in its dissolved form in marine waters. Recent publications have theorized that uncharacterized humic ligands may play an essential role in the marine iron cycle, and as such, further investigation is needed.1
To further investigate the specific origin and influence of these ligands on aquatic iron cycling, the ligands need to be extracted from the ocean’s complex matrix, isolated from the rest of the DOM. A method for this extraction was developed using immobilized metal affinity chromatography (IMAC). The method works by using a column containing
Sepharose, a cross-linked beaded form of agarose. This Sepharose acts as a chelator al-lowing us to charge the column with iron. By charging the column, the beads bind to the metal forming a coordination complex with available active sites. The iron binding lig-ands in our sample then bind to these available coordination sites, while other compounds pass through the column. A series of different eluants is then used to elute the retained ligands from the column based on different structural and chemical properties. The eluted ligands are then collected in fractions for further analysis. The optimized IMAC method results show three distinct regions of ligand classes of varying binding strength and structural composition.
To validate the method and further explore the chemistry of binding, solutions of known ligands were used as validation samples. These specific ligands were chosen due to their likely presence in aquatic environments. By testing these ligands using the optimized method, we could further infer how the humic-iron ligands interact with our iron-charged column. The results of the model ligand experiments suggest that binding to the column was typically bidentate, with two coordination sites needed for retention to occur. After testing and validation, the method was implemented for riverine, coastal, and open ocean samples. Comparison of the behaviour of natural water samples with the model lig-ands reveals potential binding characteristics and origins of HS ligands
Playing in the hero's shadow: patriarchal and neoliberal complicity in video game hero narratives
No full textIn contemporary society, many of our daily lives are strained by patriarchal oppression and neoliberal precarity. When we are disempowered like this in real life, we often turn in our leisure to entertainment that feels empowering instead. Video games regularly offer consumers such revitalizing play. However, different types of player-avatars offer different types of empowerment. The hero archetype specifically is a popular avatar type, often used in game design to inspire players. Yet the type of power play the hero offers can reinforce rather than disrupt patriarchal and neoliberal capitalist power dynamics.
My research questions how the hero archetype in video games bolsters us through, rather than against, marginalizing social hierarchies. Patriarchal and neoliberal capitalist ideology proliferate through the hero archetype’s reliance on social dualisms, which infer that one person or group (represented by the hero) is inherently superior to another person or group (represented by non-heroic entities) and so justifies the hero’s domination of these lesser “Others.” Such binary social structures perpetuate damaging concepts such as male superiority (over people who are not biologically male), white superiority (over people of colour), and elite superiority (over lower classes/the poor).
While extensive scholarship has been conducted on the presence of neoliberal and patriarchal ideology in video games, my research highlights how the hero narrative itself functions in video games to support oppressive ideologies. I argue that equality cannot exist in the hero narrative; everyone else can be equal, but the hero can only be venerated as “hero” if it stands above the rest, alone. I make this argument through an examination of hero avatar subcategories that regularly appear across video games (e.g., the epic fantasy hero in God of War and The Elder Scrolls; the war hero in Call of Duty and Battlefield; the female action-adventure
hero in Tomb Raider and The Last of Us). In targeting the hero archetype and its narrative operations in video games, I establish a model for heroic gameplay critique, which will aid future analysis of power dynamics in games as well as assist game developers and players in confronting issues of in-game marginalization.Includes bibliographical references (pages 293-338
Classical groups and self-dual binary codes
No full textSuppose that V is a symplectic space, that is, a finite-dimensional vector space endowed with
a nondegenerate alternating bilinear form. A subspace L of V is said to be Lagrangian if L
coincides with its orthogonal complement. This thesis aims to construct a simple algorithm
to compute the Lagrangians of F²ⁿ₂
as a vector space over the field F₂ up to a permutation of
coordinates. There will first, however, need to be a discussion of the classical linear groups to
achieve such a goal. In particular, we will include a discussion of the symplectic groups.Includes bibliographical references (pages 43-44
Brownian motion with velocity-dependent friction in a periodic potential
No full textThis thesis investigates the stochastic dynamics of a Brownian particle in a
one-dimensional periodic potential under velocity-dependent friction. Motivated
by physical systems such as atomic diffusion on crystal surfaces and biological
transport in structured environments, we explore how different forms of damping
influence the particle’s motion, particularly its flight-length statistics and timedependent
diffusion behavior.
We consider three friction models: Coulomb (α = 0.5), Lorentzian (α = 1), and
Gaussian damping. Each model introduces a distinct velocity dependence in the
friction coefficient, which in turn affects the particle’s ability to escape potential
wells and traverse the periodic landscape. The Langevin equation governing the
particle’s dynamics is solved numerically using a second-order accurate Milsteintype
integration scheme, adapted for multiplicative Gaussian white noise and implemented
with spline-based interpolation for computational efficiency.
Flight-length distributions are extracted from coarse-grained trajectories and
analyzed across a range of temperatures. We find that the distributions follow a
modified power-law with an exponential cutoff:
P(l) = Al⁻ᴮ exp[−C(T) l],
where the exponent is largely insensitive to temperature but increases with the
steepness of the damping function, and the cutoff parameter () decreases with
temperature following a power-law scaling. Gaussian damping yields the longest
flights and the heaviest tails, indicating reduced energy dissipation at high velocities.
The time-dependent diffusion coefficient () is computed by ensemble averaging
over stochastic realizations. For Coulomb and Lorentzian friction, () saturates
at long times, consistent with normal diffusion. In contrast, Gaussian damping
leads to persistent growth in (), revealing superdiffusive behavior even in
the presence of a confining potential.
Overall, this work provides new insights into how velocity-dependent damping
shapes transport in periodic systems. The numerical framework developed
here offers a tool for simulating non-equilibrium dynamics and can be extended to
higher dimensions, interacting particles, and experimentally relevant conditions
Improving the performance of machine learning algorithms using conceptual models: a case study of auto insurance
No full textThe integration of domain knowledge into machine learning models has been proposed as a
means to address the limitations of purely data-driven approaches. Traditional machine
learning techniques often rely on pre-defined, fixed data structures, which can overlook
valuable context-specific insights that domain knowledge provides. This study investigates the
impact of incorporating domain knowledge into the preprocessing and feature engineering
stages of machine learning models, specifically focusing on decision tree algorithms and
Support Vector Machines (SVM) within the insurance sector.
To evaluate the effectiveness of this integration, this study compares the performance of
models trained on a pre-defined dataset (A) with models trained on the same dataset after it
was enhanced with domain-specific knowledge (Revised A). The results demonstrate that the
integration of domain-specific guidelines into the feature engineering process significantly
improved the accuracy and reliability of the predictive models, particularly in complex
scenarios such as predicting customer profitability.
In scenarios where domain knowledge played a crucial role in refining features that capture
relationships within the insurance data, the enhanced models outperformed the original ones.
Conversely, for tasks where the domain knowledge had less influence, the performance
improvement was marginal. These findings suggest that integrating domain knowledge into
machine learning processes can provide a meaningful boost in model effectiveness, but the
benefits are context-dependent.Includes bibliographical references (pages 54-60
The moral implications of meconium testing technologies
No full textMeconium testing is an approach to detecting prenatal alcohol exposure. It has been
characterized by some as a promising tool for diagnosis and inciting early interventions for
people with FASD. Meconium testing inevitably reveals information about both a
newborn and their gestational parent�s health. I argue that a clinical meconium testing
practice risks raising several harms that are morally unjustified and would need to be
addressed before the implementation of the practice. I present three ways in which
meconium might risk causing or exacerbating harms to gestational parents. First, clinical
meconium testing practice may risk exacerbating existing inequalities in society by placing
disproportionate harm on marginalized people. This potential harm may occur due to the
existence of healthcare provider biases and avoidance behavior on the part of gestational
parents. Second, given the lack of accessible mental health and addictions healthcare in
Canada, there is a risk gestational parents are diagnosed with an alcohol use disorder
without having access to subsequent addictions treatment. Third, a routine meconium
testing practice risks undermining the autonomy of gestational parents and the therapeutic
relationship between them and their clinician. I offer preliminary solutions to these issues
and possible directions for future multidisciplinary study
Optimizing underwater robotic technology using artificial intelligence
No full textThis thesis proposes to optimize the navigation of autonomous underwater vehicles by enhancing
underwater acoustic communication and target sensing in complex oceanic environments.
Current deep learning techniques in underwater acoustic communication often fail to
account for domain knowledge, such as first-principles signal processing and acoustic propagation
laws, which govern chaotic underwater environments. The main focus is on the integration
of domain knowledge of the underwater environment in training neural networks. Such a
context-aware deep learning model employs theory-trained neural network to accurately learn
a non-linear map between input and output data. Details of the theory-trained neural network
largely remains unexplored, with many open questions.
This research introduces a context-aware methodology for regularizing neural networks.
We explore three approaches to deep learning-based communication among AUVs. First, we
advance long short-term memory (LSTM) neural networks for accurate underwater target detection
that moves stealthily underwater. Second, we embed communication theory within a
convolutional neural network (CNN) model in a supervised learning framework. Third, we
test a minimally viable intelligent system that enables AUVs to communicate effectively in a
hostile underwater environment. Through these venues, we provide a foundation for a more
reliable and efficient underwater communication.Includes bibliographical references (pages 78-92
Snow crab waste-derived carbon nanofertilizers: effects of foliar and soil application on growth, yield, and phytonutrients of lettuce cultivated in a controlled environment
No full textCarbon nanofertilizers (CNFs) have gained significant attention in agriculture due to their nanoscale properties, which enhance nutrient uptake, photosynthesis, plant health, and crop yield while reducing the frequency of application and minimizing environmental pollution. This study investigated the effects of CNFs derived from snow crab waste on lettuce growth, yield, and phytonutrient composition using both foliar and soil drench applications under controlled conditions. Foliar application of CNFs demonstrated a positive impact on lettuce growth by increasing leaf number, leaf area (LA), chlorophyll content, and biomass production. It also enhanced ammonium and nitrate uptake and significantly improved the phytonutrient profile, including vitamins, phenolics, total soluble sugars, total antioxidants and protein content. The results indicated that foliar application is an effective strategy for rapid nutrient absorption, leading to improved plant metabolism and overall crop quality. Soil drench application of CNFs improved soil fertility by facilitating controlled nutrient release and reducing nutrient losses. This method positively influenced root growth, chlorophyll content, and phytonutrient accumulation, leading to sustained crop productivity. Synthetic fertilizer (NPK) resulted in higher growth and yield, though CNF treatment produced a comparable yield while enhancing minerals, total antioxidants, and total phenolic compounds. These findings highlight the potential of repurposing seafood waste into CNFs that can enhance soil fertility and plant nutrition. Future research should focus on optimizing CNF formulations for various crops, assessing long-term soil health impacts, and evaluating environmental sustainability to promote the broader adoption of CNFs in modern agriculture
No place to call home: housing stigma against previous offenders and those who experience mental health concerns
No full textThe current study examined the effect of criminal record (exoneree, releasee, no record) and mental health status (mental illness, no mental illness) on landlords' willingness to rent to individuals. I hypothesized that those with a mental illness would receive fewer responses from landlords than those without a mental illness, and both releasees and exonerees' would experience more discrimination than those without a criminal record. I also expected an interaction, whereby releasees and exonerees with a mental illness would face the greatest level of discrimination. A total of 1224 emails were sent utilizing six fake email addresses, which responded to online apartment listings across Canada posted on Kijiji. Of the 1224 email inquiries, we received 414 landlord responses (33.8%). The results showed that mental illness was the strongest predictor that a landlord would not respond (70.1%); response rates dropped to 33.1% if the tenant had a criminal record. Landlords were less likely to say "Yes" to an apartment being available (57.1%) when the prospective tenant disclosed a mental illness versus when they did not (85.0%). Taken together, these findings suggest that housing discrimination is prevalent, and that mental health status may be more impactful in landlords' decisions to rent to prospective tenants than a criminal record
Geometric and topological properties of marginally outer trapped surfaces
No full textThe modern theory of gravity was introduced by Albert Einstein in 1915. In General
Relativity there is a one-to-one relationship between geometry and gravity. Black holes
are one of the most interesting predictions of general relativity. The Schwarzschild
solution or Schwarzschild black hole is named in honor of Karl Schwarzschild, who
found this exact solution in 1915 and published it in January 1916. It was the first
exact solution of the Einstein field equations other than the trivial flat space solution.
Since then black holes have become an important as well as interesting part of GR.
In the early days of general relativity, nobody believed that black holes actually exist.
However observational evidence of their existence is now overwhelming [13, 1].
One definition of black holes which is very common is that a black hole is a region
of spacetime from which even light cannot escape. But this global definition is not
very useful for understanding the dynamics of black holes. In this thesis, we want to
answer these questions: how can we define a black hole locally? And how can that
definition be used to better understand things like black hole mergers? We begin with
the definition of a marginal outer trapped surface (MOTS) and then we will discuss
what we know about them and what is our goal for the future.Includes bibliographical references (pages 53-55