Graduate Theses & Dissertations

Pages

Genetic Networks to Investigate Structure and Connectivity of Caribou at Multiple Spatial and Temporal Scales
Understanding genetic structure, connectivity, and movement of a species iscritical to management and conservation. Genetic network approaches allow the analysis of genetic information with flexibility and few prior assumptions. In chapter one, I tested the ability of individual-based genetic networks to detect fine-scale structure and connectivity in relation to sampling efforts. My findings revealed individual-based genetic networks can detect fine-scale genetic structure of caribou when using 15 highly variable microsatellite loci. Sampling levels less than 50% of the estimated population size resulted in highly disconnected networks which did not allow for accurate structure analysis; however community detection algorithms were robust in grouping closely related individuals despite low sampling. In chapter two, I used individual-based and population-based genetic networks to investigate structure, connectivity, and movement of caribou across a large study area in Western Canada. A community detection algorithm partitioned the population-based genetic network at multiple spatial scales which uncovered patterns of hierarchical genetic structure and highlighted patterns of gene flow. The hierarchical population structure results aligned with the known distribution of different caribou Designatable Units (DUs) and additional structure was found within each DU. Furthermore, individual-based networks that were constructed with a subset of samples from the Mackenzie Mountains region of the Northwest Territories revealed patterns of long-distance movement and high connectivity across the region. Author Keywords: Biological Conservation, Caribou, Community Detection, Connectivity, Genetic Networks, Structure
Is semantics activated automatically? Evidence from the PRP paradigm
Three experiments examined whether semantics is activated automatically by testing whether Arabic digits (e.g., 4), number words (e.g., four), and non-number words (e.g., rat) activate semantics in the absence of central attention within the Psychological Refractory Period (PRP) paradigm. In all three experiments, subjects performed colour discriminations as Task 1. In Task 2, subjects performed magnitude comparisons on digits (Experiment 1) and number words (Experiment 2) and size comparisons on animal words (Experiment 3). Task overlap was controlled by varying stimulus onset asynchrony (SOA). A distance effect arose in Task 2 and yielded underadditive effects with decreasing SOA for both digits and number words, consistent with these notations activating semantics in the absence of central attention, or automatically. A distance effect also arose for animal words, but it was additive with SOA, inconsistent with non-number words activating semantics automatically. Author Keywords: Automaticity, Central attention, Dual-task, Numerical cognition, Semantics, Word recognition
Machine Learning Using Topology Signatures For Associative Memory
This thesis presents a technique to produce signatures from topologies generated by the Growing Neural Gas algorithm. The generated signatures have the following characteristics: The signature's memory footprint is smaller than the "real object" and it represents a point in the n x m multidimensional space. Signatures can be compared based on Euclidean distance and distances between signatures provide measurements of differences between models. Signatures can be associated with a concept and then be used as a learning step for a classification algorithm. The signatures are normalized and vectorized to be used in a multidimensional space clustering. Although the technique is generic in essence, it was tested by classifying alphabet and numerical handwritten characters and 2D figures obtaining a good accuracy and precision. It can be used for many other purposes related to shapes and abstract typologies classification and associative memory. Future work could incorporate other classifiers. Author Keywords: Associative memory, Character recognition, Machine learning, Neural gas, Topological signatures, Unsupervised learning
Development and Use of Passive Samplers for Monitoring Dissolved and Nanoparticulate Silver in the Aquatic Environment
Silver nanoparticles (nAg) are the largest and fastest growing class of nanomaterials, and are a concern when released into aquatic environments even at low μg L-1+). Diffusive gradient in thin films (DGT) with a thiol-modified resin were used to detect labile silver and carbon nanotubes (CNT-sampler) were used to measure nAg. Laboratory uptake experiments in lake water provided an Ag+ DGT diffusion coefficient of 3.09 x 10 -7 cm2s-1 and CNT sampling rates of 24.73, 5.63, 7.31 mL day-1, for Ag+, citrate-nAg and PVP-nAg, respectively. The optimized passive samplers were deployed in mesocosms dosed with nAg. DGT samplers provided estimated Ag+ concentrations ranging from 0.15 to 0.98 μg L-1 and CNT-samplers provided nAg concentrations that closely matched measured concentrations in water filtered at 0.22 μm. Author Keywords: ICP-MS, mesocosms, nanoparticles, nanosilver, passive sampling
role of corticosterone in breeding effort and reproductive success in tree swallows
Glucocorticoids (e.g., corticosterone (CORT)) are hypothesized to mediate decisions regarding reproductive investment during breeding, but the directionality of the relationship is not clear. The CORT-fitness hypothesis posits that high levels of CORT arise from challenging environmental conditions in which an individual will conserve resources for future reproduction or self-maintenance, and thus result in lower reproductive success (a negative relationship). In contrast, the CORT-adaptation hypothesis suggests that, during energetically demanding periods, CORT will mediate physiological or behavioural changes that result in increased reproductive investment and success (a positive relationship). Inconsistencies arise due to the various species and life-history stages studied, and the complex interactions between fitness and glucocorticoids. Using an experimental approach, I investigated the relationship between CORT and reproductive success by manipulating baseline CORT levels in female tree swallows Tachycineta bicolor prior to laying using silastic implants. Implants failed to raise CORT levels of females during either incubation or the nestling stage, and maternal treatment had no effect on indices of fitness at either stage. Using a correlative approach, partial support for the CORT-adaptation hypothesis was found: There was a positive relationship between CORT and hatching success. This only occurred when CORT was measured during incubation, when baseline CORT levels may stimulate increased reproductive effort and success. In contrast, during the nestling stage baseline CORT levels were not related to reproductive investment or success. Maternal CORT levels during incubation also did not influence nestling phenotype, although nestling stress CORT levels were higher in individuals that survived to fledging. In conclusion, CORT mediates reproductive effort and success during some breeding stages, but it is still unclear why this is the case and whether this same pattern will prevail in other contexts. Author Keywords: corticosterone, HPA axis, maternal effects, reproductive success, tree swallow
Effects of Opioids on the Development and Reproductive Capacity of Japanese Medaka, Oryzias latipes
Opioid drugs are among the microcontaminants that are discharged with domestic wastewater into the aquatic environment with the potential to affect the development and reproductive capacity of aquatic organisms. To study the effects of exposure of fish to opioid drugs, Japanese medaka (Oryzias latipes) were exposed over a full life cycle to environmentally relevant (i.e. ng/L) concentrations of two opioids, codeine and fentanyl. Acute exposure of early life stage (ELS) of medaka to codeine resulted in slightly extended times to hatch, but no increase in embryo mortalities or reduced hatching success. Chronic exposure to codeine at a nominal concentration of 25,000 ng/L over a full life cycle interfered with growth, reducing the weight of the fish at maturity. Life cycle treatments with codeine at all test concentrations resulted in a significant reduction in the number of eggs produced in reproductive trials, as well as a reduction in the numbers of mature oocytes in adult females. High treatments with codeine also increased the percentage of immature sperm cells in adult males. Life cycle exposures to codeine also resulted in reduced whole-body concentrations of several hypothalamic-pituitary-gonadal (HPG) axis hormones, including reduced levels of luteinizing hormone in male and female fish and a reduction in 11-ketotestosterone in males and 17β-estradiol in females. Fentanyl did not affect reproduction or HPG hormones at the exposure concentrations tested, but high fentanyl exposures increased the mortality rate among ELS’s of the F1 generation offspring from life cycle exposed adults. The results of this thesis contribute to the literature on the environmental impacts of microcontaminants of wastewater origin and, the potential for effects in fish exposed to opioids.Keywords: Opioids, microcontaminants, fentanyl, codeine, sexual development, reproduction. Author Keywords: codeine, fentanyl, microcontaminants, opioids, reproduction, sexual development
Incidental Take and Population Dynamics of Nesting Birds in a Red Pine (Pinus resinosa) Plantation Under Single-Tree Selection Harvesting
I determined the direct influence of single-tree selection harvesting on the daily nest survival rates and nest success of 5 focal bird species within a monotypic red pine (Pinus resinosa) plantation on the western edge of the Oak Ridges Moraine in southern Ontario, Canada. I located and monitored 290 nests during the 2012 and 2013 breeding season. I used the logistic-exposure method to evaluate the daily nest survival rates of American Robin (Turdus migratorius), Eastern Wood-pewee (Contopus virens), Ovenbird (Seiurus aurocapilla), Rose-breasted Grosbeak (Pheucticus ludovicianus), and Red-eyed Vireo (Vireo olivaceus). Only five nests were destroyed as a result of forestry activity over the study period. Neither daily nest survival rates nor nest success of these focal species were substantially affected by single-tree selection harvesting. I also monitored the impact of single-tree selection harvesting on the density and territory size of 4 of 5 focal species. Ovenbird had a significantly smaller territory size but decreased density in the harvested areas. Although not significant, Eastern Wood-pewee and Red-eyed Vireo tended to have higher densities and larger territory sizes in harvested areas, whereas Rose-breasted Grosbeak showed a mixed effect as density was higher while territory size was smaller. Single-tree selection produces minor to moderate disturbance that takes place locally over a short period of time. As a result, nests that are indirectly disturbed by nearby harvesting, felling trees and mechanical operations and are not destroyed remain and adults do not appear to abandon eggs or young from the disturbance. Habitat alteration from harvesting of the general forest structure and especially the forest floor must be minimized in order to conserve forest bird species diversity. Further research examining incidental take using various intensities of single-tree selection harvesting would provide important insight into maintaining avian and forest diversity by means of forest management. Author Keywords: daily nest survival rates, forest management, Incidental Take, nest success, red pine monotypic forest, single-tree selection harvesting
Representation Learning with Restorative Autoencoders for Transfer Learning
Deep Neural Networks (DNNs) have reached human-level performance in numerous tasks in the domain of computer vision. DNNs are efficient for both classification and the more complex task of image segmentation. These networks are typically trained on thousands of images, which are often hand-labelled by domain experts. This bottleneck creates a promising research area: training accurate segmentation networks with fewer labelled samples. This thesis explores effective methods for learning deep representations from unlabelled images. We train a Restorative Autoencoder Network (RAN) to denoise synthetically corrupted images. The weights of the RAN are then fine-tuned on a labelled dataset from the same domain for image segmentation. We use three different segmentation datasets to evaluate our methods. In our experiments, we demonstrate that through our methods, only a fraction of data is required to achieve the same accuracy as a network trained with a large labelled dataset. Author Keywords: deep learning, image segmentation, representation learning, transfer learning
Widespread changes in growth, diet and depth distribution of lake whitefish (Coregonus clupeaformis) in the Great Lakes are linked to invasive dreissenid mussels
Recent declines in growth and condition of Great Lakes' lake whitefish (Coregonus clupeaformis) have been linked to ecosystem-wide changes stemming from the invasion of dreissenid mussels. To test the influence of invasive mussels on this commercially important coregonid species, we collected archived scale samples from ten Great Lake locations and analyzed long-term changes in growth rates, delta 13C and delta 15N stable isotope ratios before and after mussel establishment. There was a decrease in pre-maturation growth after establishment in all four locations where we examined back-calculated growths. In six of the seven locations with dreissenid populations, a significant increase in delta 13C and a significant decrease in delta 15N was found. In dreissenid-absent locations of Lake Superior, we did not see changes in growth or isotope ratios indicative of a major regime shift. Observed shifts in isotopic signatures provide evidence for an increased reliance on nearshore food sources and shallower depth distribution as a result of dreissenids, which likely contributed to lowered growth of lake whitefish. Author Keywords: Diporeia, Dreissenids, food web, Great Lakes, invasive species, lake whitefish
THE PROPENSITY TOWARD EXTREMIST MIND-SET AS PREDICTED BY PERSONALITY, MOTIVATION, AND SELF-CONSTRUAL
ABSTRACT The Propensity Toward Extremist Mind-Set as Predicted by Personality, Motivation, and Self-Construal Nick Fauset Multivariate regression analyses were used to determine the effects of Personality (Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness), Motivation (External, Amotivation, Intrinsic, and Identified), and Self-Construal (Independent and Interdependent) on three domains of Extremist Mind-Set (Proviolence, Vile World, and Divine Power). Participants consisted of first year undergraduate students (209 females, 76 males) enrolled in Introductory Psychology (N=279) and/or Introductory Economics (N=7), whom participated for course credit. The Motivation measure was problematic for students to complete and this variable was dropped from the model due to missing data. Decreases in Neuroticism, Openness, Agreeableeness, and Interdependent were significantly correlated with increases in Proviolence. Decreases in Agreeableness were correlated with increases in Vile World. Decreases in Openness, and increases in Agreeableness and Interdependent were significantly correlated with increases in Divine Power. These observations provide an interesting perspective on the types of Canadian undergraduate students who are more likely to score highly on measures of Extremism. Keywords: Militant Extremist Mental Mind-Set, Extremism, Personality, Five Factor Model, Motivation, Intrinsic, Extrinsic, Self-Construal, Independent, Interdependent Author Keywords: Extremism, Militant Extremist Mental Mind-Set, Motivation, Personality, Self-Construal
Habitat use within and among roosts of chimney swifts (Chaetura pelagica)
Chimney swifts are listed as Threatened nationally and in many provinces within Canada due to rapid population declines. I examined large-scale spatial variation in the maximum size of chimney swift roosts at the northern edge of their range to identify where larger roosts occur. I used multi-sourced data collected across Ontario and Quebec between 1998 and 2013. I found that larger roosts were found at more northerly latitudes, and that very large roosts (>1000 birds) only occurred north of 45°. I also investigated fine-scale patterns of chimney swift positioning inside one of the largest roosts in Ontario. Using digitally recorded images, I calculated the angular position of swifts inside the roost relative to ambient and roost temperature. I found that swifts showed a strong preference for clinging to the south facing wall and clustered more when ambient air temperature was warmer. Thus, huddling in swifts provides additional or alternate benefits, other than serving purely to reduce costs of thermoregulation at low ambient temperatures. This research contributes to the understanding of chimney swift roosting ecology and identifies large roosting sites that should be retained for conservation. Author Keywords: chimney swift, communal roosting, conservation, group size, social thermoregulation, species-at-risk
Purification and Identification of Selenium-containing C-phycocyanin from Spirulina
Selenium is an essential trace nutrient to many organisms, yet in high concentrations it is toxic. Organic selenium is more bioavailable to aquatic biota than inorganic selenium, but is usually found in much lower concentrations. Algae are known to biotransform inorganic selenium into several organo-selenium compounds, but it is unknown whether any of these bioaccumulate in the food chain. In this study, selenium was incorporated into the methionine residues of an algal photosynthetic protein, c-phycocyanin from Spirulina spp. The extent of selenium incorporation was quantified by inductively coupled plasma-mass spectrometry (ICP-MS), and the protein was identified using electrospray mass spectrometry (ES-MS). C-phycocyanin was isolated and purified from Spirulina with a final recovery of 20-30 % of the total c-phycocyanin present. Selenomethionine replaced 92.8% ± 1.22 of the methionine residues in c-phycocyanin when grown in 2.5 ppm sodium selenite. ES-MS was used to obtain protein spectra, and pure c-phycocyanin was identified. Data of full scans provided estimated masses of both protein subunits--α-chain measured at 18,036 Da; β-chain measured at 19,250 Da--close to the theoretical masses. Protein fragmentation by collision-induced dissociation and electron capture dissociation provided approximately 52 % amino acid sequence match with c-phycocyanin from Spirulina platensis. This study demonstrates the incorporation of selenium into an algal protein, and the identification of c-phycocyanin using electrospray ionization-mass spectrometry. Author Keywords:

Pages

Search Our Digital Collections

Query

Enabled Filters

  • (-) = Master of Science

Filter Results

Date

1974 - 2024
(decades)
Specify date range: Show
Format: 2024/04/24