Students enrolled in graduate programs here at Brock University will be required to submit an electronic copy of their thesis to this repository as part of graduation requirements.

Completing this process requires only a few step

  • Login using your Brock Portal Account Information
  • Click on Add your Thesis to the Repository along the right side of the screen
  • Click 'Submit a new item to this collection' below
  • Fill in enter the required data, and upload your PDF


Once your thesis has been accepted in the Repository you will receive an email confirmation along with a link to your work

Recent Submissions

  • Training Preservice Behaviour Analyst Intervention Skills in a Virtual Reality Environment

    Owusu, Gifty; Center for Applied Disability Studies
    Preservice behaviour analysts need a wide range of professional skills and shaping is one of the critical skills they must learn. This study trained preservice behaviour analysts to acquire shaping skills in a virtual reality environment using the Portable Operant Research Teaching Lab (PORTL). To date no known study has (a) evaluated the effectiveness of shaping skills training to preservice behaviour analyst or (b) attempted to teach these skills in virtual environment format. We used an AB design across participants with three preservice behaviour analysts to learn shaping skills in a virtual reality environment using the PORTL curriculum. The shaping skills comprised creating a teaching plan, setting up for a session, delivering reinforcement, and evaluating a session. For all participants, training resulted in improvement in shaping skills. Participants also maintained the shaping skills for a minimum of two weeks. Further, the effect of the training generalized to a novel learner for all participants. Additionally, participants showed high satisfaction with shaping skills in virtual reality (VR) environment.
  • An Exploration of Canadian and Nigerian High Performance Women Wrestlers’ Authentic Leadership Development Experiences in a Male-Dominated Sport

    Adeniyi, Aminat Oluwafunmilayo; Applied Health Sciences Program
    Sport management scholars have uncovered benefits from adopting an authentic leadership style among sport coaches (Kim et al., 2020), sport administrators/athletic directors (Cotrufo, 2014), and non-profit sport organization board members (Takos et al., 2018). However, there has been no scholarly attempt to learn about the experiences of high-performance women wrestlers who might aspire to become authentic leaders. Given the ongoing disparities between men and women leaders in sport organizations, arguably more should be done to understand the experiences of (and then support) future sport leaders who are women. Thus, this research study addresses these gaps by answering the research questions: (1) What are the authentic leadership development experiences of Canadian and Nigerian high-performance women wrestlers? (2) What are the formal and informal authentic leadership development experiences of women who participate in a male-dominated sport? (3) What are the perceived strengths and weaknesses of formal and informal authentic leadership development training among Canadian and Nigerian high-performance women wrestlers? Participants (n=11) engaged in one semi-structured interview that revealed their mostly informal authentic leadership development (Luthans & Avolio, 2003) experiences as members of their respective national teams. Analysis of the transcripts (65,342 words and 188 pages) followed Braun & Clarke (2006). Findings revealed five major themes and several sub-themes. Canadian and Nigerian high-performance women wrestlers’ authentic leadership development was found to be impacted and influenced by the athletes’ background influences and parental support. Participants reported developing authentic traits (i.e., resilience, optimism, confidence, and hope) that supported their development as authentic leaders both on (and off) the mat. Participants shared their experience being bullied and body-shamed for their participation in a male-dominated sport and indicated feeling both supported by and frustrated with their national sport organization/federation.
  • Solvent-free zinc-catalysed hydroboration of esters

    Zakarina, Raikhan; Department of Chemistry
    This study presents the synthesis of a novel hemilabile bidentate amido phosphine sulfide III-10 ligand and introduces a new synthetic route to the previously reported amido/phosphine III- 12 and amido phosphine oxide III-11 analogues. The present study outlines the synthesis, isolation and characterization of zincII methyl complexes III-13 and III-14 supported by III-10 and III-12, respectively. The catalytic applications of these complexes in the hydroboration of several organic substrates, including esters, nitriles, and quinoline, have been documented. When esters were subjected to the reaction conditions involving III-14 and HBcat, it was shown that ZnMe2 is formed in situ. It was then established that ZnMe2 was an active catalyst for efficient reduction of esters, resulting in moderate to good yields.
  • The Maillard reaction in traditional method sparkling wine

    Charnock, Hannah M.; Department of Biological Sciences
    The development of “aged” aromas in sparkling wine is often an indicator of quality and is characterized by caramel and toasted qualities. Compounds responsible for these aromas are derived via the Maillard reaction (MR), a non-enzymatic condensation between sugars and amino acids capable of generating a myriad of aroma compounds in a complex reaction cascade. During the production of traditional method sparkling wines, several major aging intervals take place, including the storage of reserve base wines, after the second alcoholic fermentation, and during the storage of finished wines. The work presented in this dissertation investigated the influence of amino acid and sugar precursors, plus the potential catalytic role of metal ions, on the formation of MR-associated products (MRPs) during base wine and sparkling wine aging. Four primary research objectives were addressed. The metal ion content of commercial sparkling wines produced in Niagara was established, and to our knowledge, this study represents the first report on sparkling wine metal profiles. Differences in metal composition were identified between production methods and styles, and calcium and magnesium were confirmed to be the most abundant divalent metal ions, highlighting their candidacy for involvement in MR pathways in wine. In a subsequent study, the influence of calcium and magnesium on the formation of MRPs was assessed in modified base wine treated with varying sugar and amino acid combinations during accelerated aging at 50 degrees C. Aging duration and amino acid additions were primary drivers of variation among MRPs, with calcium and magnesium having a lesser effect. A separate set of studies aimed to identify the impact of different sugar-types in dosage, the final sugar addition during production, on the formation of both MRPs and metabolite levels in sparkling wines during cellar aging. Aging duration had a greater influence on MRP and metabolite composition compared to sugar type, demonstrating that aging conditions for sparkling wine are central to the evolution of the wine matrix. This work contributes novel information to understanding the MR in mild conditions and can inform future research focused on optimizing sparkling wine composition and aging to enhance flavour in accordance with consumer preferences.
  • The Application of Chaos Game Representations and Deep Learning for Grapevine Genetic Testing

    Vu, Andrew; Department of Computer Science
    The identification of grapevine species, cultivars, and clones associated with desired traits such as disease resistance, crop yield, crop quality, etc., is an important component of viticulture. True-to-type identification has proven to be very critical and yet very challenging for grapevine due to the existence of a large number of cultivars and clones and the historical issues of synonyms and homonyms. DNA-based identification, superior to morphology-based methods in accuracy, has been used as the standard genetic testing method, but not without shortcomings. To overcome some of the limitations of the traditional microsatellite-marker based on genetic testing, we explored a whole genome sequencing (WGS)-based approach by taking advantage of the latest next-generation sequencing technologies (NGS) for achieving the best accuracy and better availability at affordable cost. To address the challenges of the extremely high dimensional nature of the WGS data, we examined the effectiveness of using Chaos Game Representation (CGR) for representing the genome sequence data and the use of deep learning in visual analysis for species and cultivar identification. We found that CGR images provide a meaningful way of capturing patterns and motifs for use with visual analysis, with the best prediction results demonstrating a 0.990 mean balanced accuracy in classifying a subset of five species. Our preliminary research highlights the potential for CGR and deep learning as a complementary tool for WGS-based species-level and cultivar-level classification.
  • A Quest for Equity in School Mathematics in Ontario: Connecting Black Secondary Students’ School Experiences and Achievement to Principal Leadership

    Morvan, Jhonel; Department of Graduate and Undergraduate Studies in Education
    Research literature on the mathematics achievement of Black students has mainly come from the United States and focused on achievement gaps with a deficit-based discourse perpetuating racial segregation and racism against Black students. The scarcity of Canadian research on the academic success, or lack thereof, of Black students in school mathematics is staggering. This mixed methods study sought to examine the relationships between the overall school experience (OSE) and the emotional well-being (EWB) of Black secondary students and their achievement in mathematics, and how principal leadership was related to these students’ OSE and EWB. Using a rich dataset from the Toronto District School Board (TDSB) and semi-structured interviews with five principals and vice-principals, this research employed four critical frameworks (critical social theory, sociocultural theory, critical race theory, and transformative leadership) to argue against this deficit ideology and to position OSE, EWB, and principal leadership as key determinants of Black students’ mathematics achievement. The findings of the study supported that Black students were disproportionately overrepresented in the Grade 9 applied mathematics assessment of the Education Quality and Accountability Office (EQAO) and less likely than any other racial groups to meet provincial standards. A binary logistic regression confirmed that a White student at the TDSB is almost four times more likely to meet provincial standards in Grade 9 EQAO mathematics assessment than a Black student. This model also established that both the OSE and the EWB were statistically significant predictors of achievement in the Grade 9 EQAO mathematics assessment. Research participants pointed to principal leadership, curriculum implementation, relationship building, and participation and engagement, among other things, as having the potential to significantly impact Black students’ OSE, EWB, and mathematics achievement. These findings add to the limited literature on Black students’ mathematics achievement and offer a significant contribution to the conversations aiming at challenging all educators, school leaders, and policymakers to address anti-blackness sentiments and anti-Back racism in school mathematics. Lastly, implications of these different results pointed to directions for future research aiming at understanding the situation of Black students in mathematics classes.
  • Effects of cross-fostering on behaviour and neural development in Octodon degus pups

    Attlas, Gurprince; Department of Psychology
    Parental care is essential for social, behavioural, and neural development in offspring. In rodents, parental separation affects the amount of parental care and progression of offspring development. Work to date has focused on maternal and paternal deprivation, but it is unclear how cross-fostering, another form of parental-offspring instability, can affect offspring behaviour and brain development. Stress significantly suppresses neurogenesis and increases inflammation in the dentate gyrus (DG) of the hippocampus, but this can vary between sexes. Octodon degus are highly social rodents with precocial offspring that can receive care from both parents, allowing us to study the effect of early life stress on pup development. This study investigated the effect of cross-fostering on parental care, offspring behaviour and hippocampal development in female and male degus. At postnatal day 8, degus were assigned to either control (pups remained with parents and siblings), partial cross-foster (CF; one pup/litter was cross-fostered), or full CF (the entire litter was cross-fostered). At weaning (5-weeks-old), offspring brains were collected for immunohistochemistry to examine DG volume and expression of immature neurons (using doublecortin, DCX) and microglia (using Iba-1). Partial and full CF did not affect parental care provided by parents compared to controls. In offspring, play fighting behaviour was significantly higher in partial CF females compared to controls. Males initiated play fighting more often than female pups but were not affected by CF. Partial and full CF did not affect the DG volume and optical density of DCX compared to control pups. Full CF pups had fewer ameboid microglia in the dorsal DG compared to controls. In the ventral DG, full CF pups had more intermediate microglia than controls. Our study indicates partial CF affects play fighting behaviour in females, while full CF affects microglia morphology in both sexes suggesting potential changes in hippocampal inflammation and plasticity. This indicates that CF category affects females and males differently depending on the endpoint measured and that these effects do not seem to be associated with changes in parental care. This study contributes to our understanding of how early life environments affect offspring behaviour and brain development in both sexes
  • The Effects of Strength Spotting on Self-Reported Parenting Competence and Parent-Child Relationship Quality in Parents of Autistic Youth

    Yu, Kevin Han Xiang; Center for Applied Disability Studies
    Autism research has often focused on the deficits and challenges experienced by younger children on the spectrum, resulting in the underrepresentation of autistic youth and adolescent populations. Moreover, the emphasis on deficits may negatively impact the quality of parent-child relationships and self-perceived parenting abilities among parents of autistic children. Alternatively, strengths-based interventions such as Strength Spotting may facilitate a shift toward recognizing the positive characteristics of autistic people and benefit families with children on the spectrum. The present pilot study sought to explore the effects of Strength Spotting on parent-child relationship quality and self-reported parenting competence among parents of autistic youth. Nine parents participated in the study, where they learned about and implemented Strength Spotting with their autistic children. We used a pre-post-test design to measure parent-child relationship quality and parenting competence changes. The results revealed non-significant differences in relationship quality, with moderate to large effect sizes suggesting potential improvements. In contrast, parenting competence significantly improved with large effects when comparing pre- and post-intervention stages. Despite the non-significant changes in relationship quality, this study demonstrated high social validity and, to our knowledge, was the first to apply Strength Spotting with parents of youth on the autism spectrum. These preliminary findings hold promise for autistic people and their families, highlighting the importance of recognizing positive behaviours among autistic family members. Furthermore, positively oriented interventions such as Strength Spotting may also have beneficial personal and clinical implications for autistic people and their families.
  • Advancing Resilience and Equity in Canadian Municipal Climate Adaptation

    Lepp, Madison; Environmental Sustainability Research Centre
    Effective climate adaptation must build resilience and advance social equity. Most municipalities now recognize the importance of furthering resilience and equity in their climate adaptation planning. To date, however, there has been limited attention to whether municipalities are currently incorporating resilience and equity into their planning and how to assess status and track progress towards resilience and equity in climate adaptation. Accordingly, this research responds to the need to better understand how principles of resilience and equity are articulated in Canadian municipal climate adaptation plans (Study One) and to develop a framework to strengthen resilience and equity in municipal climate adaptation (Study Two). Study One evaluates ten Canadian municipal adaptation plans through a content analysis. Plans were evaluated using a coding protocol consisting of 26 indicators based on 10 principles of resilience and equity. The analysis revealed three key findings that are important for policy and practice: (1) Canadian municipal adaptation plans prioritize resilience over equity, (2) complex theories of resilience were less commonly operationalized, and (3) distributional equity is insufficiently operationalized in Canadian municipal adaptation plans. Study Two conducts a literature review and survey with municipal practitioners (n=15) to develop an index to assess resilience and equity in municipal climate adaptation planning. The Climate Resilience and Equity (CRE) Index is comprised of 10 principles and 26 indicators. The index is a first step in advancing an approach to strengthen our understanding of how indicators of resilience and equity are integrated into municipal adaptation planning. The CRE Index could be used by researchers and practitioners to mainstream resilience and equity in municipal climate adaptation planning and decision-making. Overall findings have many implications for theory and practice including, but not limited to, improved climate adaptation, enhanced community well-being, and the fostering of more inclusive and sustainable urban development. This thesis highlights the value of centring resilience and equity in adaptation planning, emphasizing that a comprehensive approach not only bolsters municipalities' capacity to navigate climate challenges but also contributes to broader societal goals, ensuring that no community is left vulnerable in the face of environmental change.
  • Paper Parks or Protection: Evaluating Atlantic Canada's Marine Protected Areas

    McIntyre, Sydney; Environmental Sustainability Research Centre
    Reversing biodiversity loss is one of society’s most pressing challenges. In response, marine protected areas (MPAs) are arguably one of the most effective conservation solutions. Yet, the outcomes of MPAs are highly variable. Some deliver positive biodiversity outcomes while others are criticized for being “paper parks”; a term used to describe protected areas that are designated on paper but offer little contributions towards the conservation of nature. The current protection levels of Canada’s MPAs are largely unknown. In this major research paper (MRP), I evaluated the protection levels for eight MPAs located in Canada’s Atlantic Ocean. The analysis revealed that over half (62.5%) of Atlantic MPAs are incompatible with conservation due to the heavy presence of offshore oil, fishing, and shipping. These results suggest that enhancing the levels of protection in the MPAs on Canada’s east coast is required for MPAs to contribute effectively to biodiversity conservation and human well-being.
  • Investigating Individual Differences in the Aftereffects of Self-Control Exertion

    Stirpe, Jacob; Department of Psychology
    The aftereffects of self-control exertion have been debated by psychologists over the last two decades. Among those who claim there are aftereffects of self-control exertion, some contend that self-control acts as a limited resource that depletes as you use it (Baumeister, Heatherton, & Tice, 1994), while others suggest that exerting self-control provokes a change in attention and motivation from ‘have-to’ goals to ‘want-to’ goals (Inzlicht & Schmeichel, 2012). Main effects of self-control exertion have been found that support both of these theories, but so have many null results. Individual difference models have rarely been applied to these theories despite the fact that they could potentially account for the mixed pattern of results. Indeed, individual differences in trait approach motivation have been found to moderate the aftereffects of self-control on emotionally neutral cognition tasks but have yet to be investigated in tasks with motivationally salient, approach-related goals or stimuli. The current study looked to investigate how self-control exertion will affect subsequent behaviour in approach-based tasks and whether or not this behaviour is moderated by individual differences in trait approach motivation and/or value driven attention. Across two studies, participants reported trait levels of approach motivation and value-driven attention, were assigned to exert high or low levels of self-control, and were then presented with a gambling task (Study 1) or an RSVP image detection task with low and high approach-motivated images (Study 2). The results did not show evidence of a main effect of self-control exertion in either study, but Study 2 showed some evidence of individual differences in trait approach motivation and self-control exertion interacting to modulate attention to approach-motivated stimuli. Specifically, participants who had just exerted high levels of self-control and were low in trait reward responsiveness showed a greater effect of approach motivated stimuli than those low in reward responsiveness and/or those who did not just exert self-control. The present results fail to provide support for either competing self-control theory but suggest that individual differences can play a significant role in the aftereffects of self-control exertion.
  • Topic Modeling-based Logging Suggestions for Java Software Systems

    Akter, Mehenika; Department of Computer Science
    Log statements help software developers and end users get informed about different valuable run-time information while log levels categorize the severity of that information. Researchers have been working extensively on log-related problems for the last two decades. As a result, a good amount of research has been conducted on logging and its practices. However, determining which topics can be logged from a system has a potential to work on. To implement our study, first, we examined the code snippets from some renowned open-source Java language-based projects. We collected the logged methods from nine applications and after preprocessing the methods and extracting our required data, we applied some renowned topic models: Latent Dirichlet Allocation (LDA), Latent Semantic Analysis (LSA), and Non-negative Matrix Factorization (NMF). In the first part of the results, we showed how the topics are related to logging to investigate the alignment between topic modeling and logging. Our dataset, enriched with meaningful words related to method functionality, is subjected to LDA analysis. Results indicate that topics with the highest sum of word probabilities are more likely to be logged. In the second section, we listed the popular topics with their associated words from different systems generated by LDA. In the last part of the results, a comprehensive result was shown by evaluating the performance of the models using coherence scores. We believe that our research will not only be useful for its result and evaluation but also be helpful for future researchers by providing a unique dataset.
  • The Susceptibility of Medically Important Mosquito Species to Insect-Specific Virus Infection

    Williams, Amanda; Centre for Biotechnology
    Arthropod-borne viruses (arboviruses) cause disease in humans and animals throughout the world. Many factors exacerbate the risk of virus transmission over time, including climate change and global transport. Developing methods to control their transmission is of interest, and insect-specific viruses (ISVs) pose a novel avenue for biocontrol of arboviruses at the vectorial level. Cell-fusing agent virus (CFAV) (Flaviviridae: Flavivirus) is an insect-specific flavivirus that has been seen to influence the replication of arboviruses in vitro and in vivo. The maintenance of this virus within a persistently infected colony of Aedes aegypti was investigated. Between 2022 and 2023, the colony infection rate decreased. The capacity of the virus to be horizontally transmitted through cohabitation was also assessed with Ae. aegypti, by housing infected mosquitoes with uninfected mosquitoes for 7 days. No horizontal transmission was observed. The use of CFAV-infectious cell culture to infect Ae. aegypti larvae was also tested. Infection was successful in low proportions. These results gave insight into CFAV transmission pathways and raised questions about the factors that can affect stability of CFAV over time. Negev virus (NEGV) and Piura virus (PIUV) are part of a recently discovered genus called Nelorpivirus. Not much is known about their transmission or their effects on arboviruses. Here, Ae. aegypti and Culex pipiens larvae were successfully infected via infectious cell culture. There were higher infection rates in Ae. aegypti. Both species were able to vertically transmit PIUV while only Cx. pipiens could transmit NEGV to offspring. These results demonstrate that both medically important mosquito species are susceptible to infection by NEGV and PIUV. This is also the first experimental evidence of their capacity for vertical transmission.
  • LSTM-oriented Handover Decision-making with SVR-based Mobility Prediction in 5G Vehicular Networks

    Chowdhury, Shajib; Department of Computer Science
    The advancement of 5G technology is initiating a transformation era for Vehicular Networks (VN), enabling seamless communication among vehicles and other entities. Connected vehicles hold significant potential for improving traffic safety, and enhancing in-vehicle entertainment. With the increasing of vehicular applications, the necessity for reliable, high-bandwidth, and low-latency connections has become increasingly paramount. Ensuring consistent connections in dynamic vehicular settings remains an ongoing challenge, especially given the necessity for smooth Handovers (HO) between transmission points as vehicles move rapidly. Frequent handovers, due to the limitations of communication range, can impact user experiences, especially in safety-critical situations. One potential solution involves transitioning to network virtualization to address the challenges posed by ultra-dense networks and the limited communication range in 5G. To tackle these challenges, we present an approach based on mobility prediction for selecting virtual cells using Support Vector Regression (SVR) and making Handover (HO) decisions using Long Short-Term Memory (LSTM). Our method, named M-LSVR, focuses on forming user-centric virtual cells based on network attributes and real-time data. The dynamic adjustment of virtual cell size using predictive mobility ensures stability and reduces unnecessary handovers. Integrating mobility prediction with HO decision-making aims to establish a more stable connection, enhancing the quality of virtual cells in high-mobility vehicular environments. This approach aims to optimize the user experience by minimizing unnecessary tower switches and creating efficient, high-quality virtual cells in the 5G vehicular network.
  • Athletes vs. Coaches: Perspectives on Social Media

    Gorrell, Elyse; Applied Health Sciences Program
    This dissertation’s purpose was to understand the different perceptions of social media in athletes and coaches. Previous research has not given a thorough examination of social media’s effects on coaches, or the consequences of social media for meaningful relations between coach and athlete. Previous literature suggests that not only do athletes not understand the influences or implications of social media, but also that these influences and implications could alter an athlete’s mentality for performance. Previous research also recognized that athletes’ preoccupation with social media is a perceived challenge for coaches. Goffman’s (1959) Presentation of Self in Everyday Life and Sissela Bok’s (1978) Lying functioned as significant theoretical frameworks within this dissertation which was also guided by an existential phenomenological orientation to embodied experience. Methods used included 1) semi-structured interviews with 10 high-performance competitive athletes, from individual sports; 2) content analysis of 13 post submissions by athletes with an athlete reflection; and 3) semi-structured interviews with 6 high-performance competitive coaches. A phenomenological analysis sequence was applied to the data sets, which consisted of parsing the responses to the interview questions into units or phrases of meaning, maintaining horizontalization, and creating a rich description that represented the meanings discerned from the data analysis. Imaginative Free Variation (IFV), guided by the lifeworld existentials of Body, Space, Time and Relation (bstr), was also a strategy used to deepen the examination of themes and summary descriptions. A manifest and latent analysis was utilized on the posts (“phase 2”) to describe the post while suggesting a plausible interpretation that the athlete’s reflection could be compared against. The posts were also used within the coach interviews in order to gain the coaches’ perspectives of the selected posts. The study aimed to bridge the gap in understanding the relation between coaches and athletes and how the effects that social media has on athletes and their performance. The findings suggest that athletes are aware that people are viewing them, and that they are having a difficult time regulating information that they see on social media or how they are using social media. The findings from the “phase 2” analysis suggest that athletes believe that they have insider meaning that the average viewer does not when viewing certain social media posts. These findings also suggest that the coach-athlete relationship runs the risk of becoming less effective, because there is less value given to interactions within the physical environment.
  • Generating Models of Human Gait in Patients with Parkinson’s Disease

    Navikevicius, Tristan; Department of Computer Science
    Parkinson’s disease is an extremely debilitating condition where the brain is not producing enough dopamine to accurately coordinate movement. One symptom of Parkinson’s disease, freezing of gait, prevents the affected person from either starting to walk or continuing walking. It usually begins in the advanced stages of the disease. The primary medication for Parkinson’s disease, Levodopa, is only partially effective for the treatment of freezing of gait. The dataset studied in this thesis provides time-series gait data of individuals’ gait while performing four different tasks, each having increased complexity over the previous ones. This thesis looks at a time-series gait dataset and performs symbolic regression through genetic programming on that dataset to predict fall likelihood and to create models of the gait of people with and without Parkinson’s disease including people who may be experiencing freezing of gait while factoring in their medication status (ON or OFF). The fall prediction experiment suggests that the GP models can predict the likelihood of falling based on the individual’s gait. The models provide insights into how Parkinson’s disease and freezing of gait impact gait patterns in people who have the disease vs. those who do not and enables us to compare the gait of individuals in different groups. It was found that, as expected, gait was similar within groups and different between groups. We also found that for some individuals it was not possible to distinguish between ON and OFF medication states. Future work might include determining the best models for each individual or group, attempting to find a model that accurately represents the individual or group rather than the individual trials.
  • Attention-Based Generative Model in Deep Evolutionary Learning: A Many-Objective Approach to Multi-Target SMILES Fragment-Based Drug Design for Cancer

    Ahmed, Madiha; Department of Mathematics
    Cancer remains a global health challenge, necessitating novel drug discovery methods. This graduate thesis introduces two computational frameworks for multi-target drug design in cancer therapy, firstly, by integrating Deep Evolutionary Learning (DEL) with a Transformer-based model. Departing from the traditional use of Variational Autoencoder (VAE), this research employs a Transformer-based generative model, capitalizing on its superior ability to capture long-range dependencies within molecular sequences to develop an understanding of the complex molecular grammar. Secondly, the research further evaluates the efficacy of a more granular fragmentation method than the one originally employed in DEL. These two proposed modifications of DEL: (i) Transformer-based model integrated in the original DEL framework and (ii) a fragmentation technique in finer granularity incorporated in the original DEL framework, are each evaluated and compared against the original DEL framework, the benchmark, in their molecular generative capabilities of targeting multiple biomarkers in cancer progression. In essence, the Transformer’s parallel processing capabilities enhance the drug design efficiency in terms of enhancing the diversity of novel and valid population samples produced and generating the highest-ranked novel molecule with the most optimal set of protein-ligand binding affinities. By optimizing the fragmentation technique, it is observed that it also performs well in maintaining a high novelty and validity of molecular compounds and interestingly, in drug design tasks involving specification of the off-targets, it produces a higher number of novel compounds that satisfy the objective thresholds compared to the benchmark. Overall, we believe that these are two approaches that can be explored for developing cancer treatments, and can also offer potential solutions for other diseases requiring multitarget interventions.
  • Environmental and social influences of foraging behaviour and maternal investment with a note on the mating frequency of the eastern carpenter bee, Xylocopa virginica

    Duff, Lyndon; Department of Biological Sciences
    This thesis studies how weather and sociality influence the foraging effort input costs of brood production, and that multiple mating contributes to low-group relatedness and non-kin sociality in the Eastern carpenter bee, Xylocopa virginica. In Chapter 2, using a natural experiment which included a drought and three normal weather years, we found that mothers foraged more in a drought but had smaller offspring than most normal weather years. Droughts conditions lead to fewer flowers, and it follows that less abundant food resources should lead to a trade-off of more work for fewer resources. Indeed, we found that during the drought, the size of pollen loads that mothers carried were smaller than in every other normal year showcasing the impact of the drought. In Chapter 3, I assess the foraging effort in both solitary and social foundresses showing that the foraging input costs are much higher for social primary foundresses than for solitary foundresses. That foraging costs are higher for social foundresses is atypical compared to most literature, which usually suggests that solitary behaviour is more costly than social behaviour. Lower social costs implies that social behaviour should become the more predominant phenotype in the population, however, in our Niagara population there was always a mix of solitary and social phenotypes. A mix of solitary and social phenotypes suggests balanced selection for both phenotypes. That the social phenotype is more costly is interesting and we suggest that social primary foundresses must feed their nestmates throughout the foraging period. Nestmate feeding has been documented in this bee before, so it seems like the most plausible explanation. In Chapter 4, we assessed the mating frequency of female carpenter bees using behavioural observations and microsatellite markers. We found using both sets of data that X. virginica mates multiply. Multiple mating has direct implications on the genetic relatedness of already low relatedness in non-kin social groups. Non-kin sociality is understudied, and we hope that these findings spur additional studies.
  • Reinforcement Learning-based Time-Dependable Modelling of Fog Connectivity for Software-Defined Vehicular Networks

    Ferdous, Jannatul; Department of Computer Science
    Connected vehicles are crucial in strengthening vehicular and Intelligent Transport Systems (ITS) by enabling autonomous and dynamic data sharing across the vehicular network. Extensive research has been conducted to predict connectivity, alongside thedevelopment of diverse techniques to manage this essential aspect. In recent times, learning methodologies have become increasingly popular for their ability to effec-tively handle sophisticated models adaptively. Various machine learning algorithms have been demonstrated as convincing methods for rendering any system flexible andpredictive. We thus propose a Learning based Adaptive Connectivity Estimation Model LACM. This model calculates and enhances the connectivity among differentstates and actions, monitoring their changes over time. The purpose of this model is to accurately depict the current connectivity status and predict potential fluctuations in fog connectivity. This model will utilize networking and vehicular characteristics to make the accuracy of its predictions. The design of this model aims to tackle the complexity of the problem by incorporating detailed data into a large state space representation, thereby enhancing adaptability. The second part of our work proposes a Time Dependent Connectivity Estimation Model, TDCM. Incorporating time dependency in the model helps to forecast the alterations in cluster lifestyles. It shows the progression of cluster evolution, significantly contributing towards achieving a stable and reliable network. Utilizing Long Short-Term Memory within an RL-based framework enables the system to enhance decision-making accuracy through predictions related to connectivity and network maintenance. Extensive analysis conducted through realistic simulations demonstrated that both LACM and TDCM strongly support estimating and maintaining stable connectivity over time. Our evaluation compared a previous state-of-the-art approach, showing that LACM and TDCM consistently enhanced the connectivity within the network.
  • Combining the Power of Attention Models and Many-objective Computational Intelligence Algorithms for Drug Design

    Aksamit, Nicholas; Department of Computer Science
    AI-based approaches have been recently applied to in silico drug design. However, existing approaches and protocols consider the absorption, distribution, metabolism, excretion, and toxicity (ADMET) pharmacokinetic properties of drug candidates in a later stage of drug design processes, where failure is most costly. To address this challenge, this research work aims to achieve three objectives. First, it explores the use of Transformer-based models for ADMET prediction based on a hybrid fragment-SMILES tokenization scheme and two training strategies. Second, it evaluates the performance of contrastive Transformer-based latent models for molecular generation. Third, it applies many-objective computational intelligence algorithms in the continuous latent space generated by a Transformer model to generate optimal drug candidates that fulfill ADMET and other essential properties in parallel. The results of this research work demonstrate superiority in the hybrid approach over SMILES in predicting ADMET properties. Furthermore, the system proposed in this study integrates metaheuristics with ADMET prediction and latent Transformer models for solving a drug design problem. A comparative study shows effectiveness of computational intelligence towards a many-objective drug design problem, where 1718 drug-like molecules are obtained after application of a strict filtering criteria.

View more