Chelsea Aristone

Consultants: Dr. Sigal Balshine, Hossein Mehdi

Impact of Wastewater Effluent on Zooplankton Population Dynamics in Winter Conditions

Zooplankton exist in most fresh and marine bodies and play an important role in food webs by controlling phytoplankton populations and acting as a food source for invertebrates and fish. Analyzing their populations can provide insight into the productivity of an area, as well as the dynamics of the community. Zooplankton can provide insight on food web dynamics from the perspective of bottom-up control on invertebrates, fish, and other larger organisms that consume zooplankton. This project aimed to determine how the density, richness, and diversity of zooplankton populations change between sites in the wastewater gradients of Hamilton Harbour during the winter.

To determine the influence of the wastewater, samples were taken from two different wastewater treatment plants (WWTPs) in Hamilton Harbour: Dundas and Woodward. At each of the WWTPs samples were taken from two reference sites and three sites at different points in the wastewater gradient. These samples were preserved and then later analyzed by counting the number of individuals of each species for each sample that was taken.

Results from the samples taken indicate that the Dundas WWTP had a higher abundance of zooplankton in the two most contaminated sites than two of the cleaner sites. The Woodward WWTP had a higher Menhinick Richness at the two cleaner sites than second and third most contaminated sites.

This research may be furthered by studying the trends at these same sites during the summer months to see the impact of seasonality on zooplankton communities near the WWTP.

Christy Au-Yeung

Consultants: Dr. Margaret McKinnon, Dr. Sue Becker (Professor- Department of Psychology, Neuroscience and Behaviour)

Effectiveness and neurophysiological changes associated with neurofeedback-mindfulness Intervention in Post-Traumatic Stress (PTSD) with and without dissociation

Current Post traumatic stress disorder (PTSD) treatments have high dropout and nonresponse rates, and are unable to meet the unique needs of a certain a subtype of PTSD who experience dissociation (PTSD+DS). A promising, yet unexplored intervention is neurofeedback-mindfulness which provides direct feedback of brain activity to participants while they are performing mindfulness activities. It is hypothesized that it can lead to positive neurophysiological and behavioural improvements due to neuroplastic changes in top-down attentional control circuits in the brain. Certain aspects are hypothesized to address dissociative symptoms. Participants from the PTSD inpatient unit at Homewood Healthcare Centre chose to either be placed in the treatment as usual group (TAU) or treatment as usual with mindfulness-neurofeedback sessions (TAU+MTP).  The neurofeedback sessions were completed up to three times a week for a duration of four weeks using the InteraXon MuseTM headband. Participants in the TAU+MTP group also completed weekly questionnaires on mood, PTSD symptoms and dissoiative symptoms. Both groups completed pre- and post-testing, consisting of psychological tests and an optional EEG session. The EEG session comprised of a five minute resting phase and alternating working memory and autobiographical memory recall tasks. MTP related behavioural and neurophysiological changes were assessed by comparing the measures of brain activity derived from EEG such as percentage of correct network activation and percentage of correct network switches, and scores from the psychological testing and questionnaires. While full results are pending, we anticipate positive behavioural changes in TAU+MTP group, reflected in changes of neurophysiology and brain activation patterns.

Relevant Sources: (Harvard Anglia Ruskin)

Boyd, J.E., Lanius, R.A. and McKinnon, M.C., 2018. Mindfulness-based treatments for posttraumatic stress disorder: a review of the treatment literature and neurobiological evidence. Journal of Psychiatry & Neuroscience : JPN, [online] 43(1), pp.7–25. Available at: <> [Accessed 24 Jan. 2020].

Lanius, R.A., Brand, B., Vermetten, E., Frewen, P.A. and Spiegel, D., 2012. The dissociative subtype of posttraumatic stress disorder: rationale, clinical and neurobiological evidence, and implications. Depression and Anxiety, 29(8), pp.701–708.

Sitaram, R., Ros, T., Stoeckel, L., Haller, S., Scharnowski, F., Lewis-Peacock, J., Weiskopf, N., Blefari, M.L., Rana, M., Oblak, E., Birbaumer, N. and Sulzer, J., 2017. Closed-loop brain training: the science of neurofeedback. Nature Reviews Neuroscience, [online] 18(2), pp.86–100. Available at: <> [Accessed 23 Jan. 2020].

Van Ameringen, M., Mancini, C., Patterson, B. and Boyle, M.H., 2008. Post-traumatic stress disorder in Canada. CNS neuroscience & therapeutics, 14(3), pp.171–181.

Gemma Barber

Consultant: Dr. Joyce Obeid

The Relationship Between Aerobic Fitness and Pulse Wave Velocity in Children with Juvenile Idiopathic Arthritis and Healthy Controls

Children with Juvenile Idiopathic Arthritis (JIA) may be at an increased risk of poor heart health, as demonstrated by increased arterial stiffness compared with their healthy peers. Aerobic fitness is associated with lower artery stiffening in healthy children. To date, no studies have examined the relationship between aerobic fitness and arterial stiffness in JIA. Therefore, the objectives of this study were to: (1) compare arterial stiffness in children with JIA and healthy controls (CNT), and (2) assess the relationship between aerobic fitness and arterial stiffness in these youth. Participants (7-17 years) completed two study visits. Aerobic fitness was assessed using a graded cycling test to exhaustion with continuous gas collection to measure VO2peak. Arterial stiffness was assessed with whole-body pulse wave velocity (PWV), where higher PWV indicates greater arterial stiffness. A total of 27 participants with JIA (18F, age 12.9±3.2 years) and 15 CNT (5F, age 13.8±2.4 years) were included. There were no significant differences in PWV between groups (JIA: 5.1±0.4 m/s vs. CNT: 5.3±0.5 m/s, F(1, 39)=0.518, p=0.476). Height, but not group (p=0.749) or VO2peak (p=0.606), significantly predicted PWV in a linear regression (F(3, 38)= 7.962, p<0.0005, adj. R2=0.34). Future directions include ensuring that groups are better matched for age and sex, measuring the change in PWV over time and testing children presenting with different levels of disease activity. Understanding the link between aerobic fitness and arterial stiffness may help identify a potential target (aerobic fitness) for intervention to improve or maintain heart health for children with JIA.

Paula Bosca

Consultant: Dr. Laura Parker

Central Galaxy Star Formation in X-Ray Bright and X-Ray Faint Galaxy Clusters

Galaxy clusters are giant structures composed of up to several thousand galaxies bound together by their mutual gravitation. At the gravitational center of most clusters is a massive galaxy. Due to their location, material accretes directly onto these central galaxies unlike orbiting satellite galaxies. The properties of these massive central galaxies can shed light on the evolution of galaxies in clusters and the growth of the most massive galaxies in the Universe.

Clusters are permeated with hot gas that can be detected through their X-ray emissions. The X-ray luminosity can be used as a proxy for the dynamical state of a cluster as more relaxed clusters have higher X-ray luminosity. Previous studies have shown that satellite galaxies in clusters display significantly different star formation rates in X-ray strong versus X-ray weak clusters and indicate that ‘quenching’ processes are more efficient in X-ray strong systems. In this study, we examine the relation between star formation and X-ray luminosity of central galaxies in the Sloan Digital Sky Survey Data Release 7 (SDSS-DR7). We find that the fraction of star-forming galaxies is higher in X-ray strong systems. This suggests that star formation is fueled by ongoing gas accretion, with relaxed systems showing greater star formation than their less relaxed counterparts. 

Luke Buckler

Consultant: Dr. Juliet Daniel

The role of PD-L1 in triple-negative breast cancer in women of African ancestryTriple-negative breast cancer (TNBC) is an aggressive breast cancer subtype that is difficult to treat due to a lack of expression of estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2. Intriguingly Women of African ancestry (WAA) are diagnosed with TNBC at higher rates than Caucasian women and also experience poorer clinical outcomes. The genetic factors driving this racial disparity represent a gap in the literature surrounding this disease, however increasing evidence suggest a link with inflammatory pathways. The objective of this study was to correlate the expression of the immune marker PD-L1 on the incidence and survival rates of TNBC in WAA. We employed immunohistochemistry (IHC) to investigate the expression and subcellular localization of PD-L1 on a tissue microarray consisting of TNBC samples from WAA and Caucasian women.  Expression profiling of each sample was quantified using QuPath quantification software and resulting scores were correlated with respective clinical data. The Allele Frequency Calculator from Ensemble was used to analyze population genetics surrounding variants in the PD-L1 gene. We found that the CD274 gene, which encodes PD-L1, has more variants with higher frequencies in WAA than in Caucasian women, indicating that PD-L1 mutations are more common among African populations which may contribute to the poor prognosis of TNBC in WAA. The expression of PD-L1 in the TNBC tumour samples is expected to correlate to a worse clinical outcome from the patient. This project provides a foundation for future studies further looking into immune markers in TNBC in WAA.

Diego Covarrubias

Consultant: Dr. Jianping Xu

Effect of UV-induced genetic mutations on the thermoadaptability of the White-Nose Syndrome pathogen, Pseudogymnoascus destructans.

The fungus Pseudogymnoascus destructans is a psychrophilic pathogen of European origin which causes White Nose Syndrome in North American bats. First appearing in North America (NA) in 2006, it has caused population collapse in almost all hibernating bat species. Despite its psychrophilic and clonal nature, it shows extremely fast adaptability to a variety of climatic and geographic conditions. The goal of this project is to better understand mechanisms of thermotolerance in P. destructans using UV mutagenesis. The survivability of P. destructans strains (i.e., US15 the earliest known strain in NA) was tested, by plating on SGA+CI media at 102, 103, and 104 cells per mL. P. destructans will be irradiated under a USE G30T8 253.7nM germicidal lamp in a laminar flow containment hood (Forma Scientific Class II A/B3 BSC) at a distance of 15cm. Six radiation exposure times will be used for the three dilutions: 0s, 10s, 30s, 45s, 1m, 2m. After allowing the fungus to grow in a 14C incubator, a die-off curve to determine the optimal exposure time will be produced. The same methodology will be repeated for multiple Canadian and American strains that are clonal descendants from the initial point of introduction in NA. Temperature tolerance to 14C, 18C, and 20C will be tested. As P. destructans is not usually able to survive in temperatures above 20C, UV-induced mutations will call for investigation of emergent thermoadaptability. This research will expand on the limited knowledge of how this fungus is able to adapt and spread within NA.

Forsythe, A., Giglio, V., Asa, J. and Xu, J., 2018. Phenotypic Divergence along Geographic Gradients Reveals Potential for Rapid Adaptation of the White-Nose Syndrome Pathogen, Pseudogymnoascus destructans , in North America. Applied and Environmental Microbiology, 84(16), pp.e00863-18, /aem/84/16/e00863-18.atom.

Palmer, J.M., Drees, K.P., Foster, J.T. and Lindner, D.L., 2018. Extreme sensitivity to ultraviolet light in the fungal pathogen causing white-nose syndrome of bats. Nature Communications, 9(1), p.35.

Abhigyan Dwivedi

Consultant: Dr. George Dragomir

Cellular Automaton Model for Glioblastoma Cell Proliferation and Migration with an Analysis of Growth Parameters that Explain Frontal Lobe Occurrence Bias

Glioblastoma multiforme (GBM) is a highly proliferative and recurrent grade IV astrocytoma. Treatment includes a surgical resection and then a combination of chemo/radio therapy to prevent recurrence, however, this is highly ineffective. Modeling GBM and analysing parameters of growth allows for more accurate growth predictions and directly improves clinical outcomes.  Cellular automaton models have been successfully applied to many types of cancers including gliomas and have accurately predicted real growth patterns. We modify this model to make it specific for glioblastoma growth using open source clinical data. Our cellular automaton relies on simple rules that govern the fate of each surrounding neighbor cell by previously encoded probabilities based on the extrapolated clinical data. The model is highly suitable for glioblastoma since it allows for the existence of a necrotic core, a quiescent middle portion and a proliferative covering. Furthermore, it can be easily limited to exhibit a Gompertzian sigmoidal growth pattern commonly observed in glioblastoma growth to mimic real world conditions. Glioblastoma like many cancers is dependent upon certain parameters. We explore O2 partial pressure, vascularity which determines nutrient availability, and cell death rate. High oxygen and nutrient availability are responsible for the initial exponential growth of tumor cells. As resources wain, and cell death rate increases the growth asymptotes. We analyze these parameters independently and interdependently to show increased nutrient and oxygen availability explains the GBM occurrence bias in the frontal lobe and helps model the GBM.

Lucas Eckert

Consultant: Dr. Sigal Balshine

Co-evolution of Reproductive Anatomy and Behaviour across Fishes

Fishes, one of the most specious group of vertebrates, display incredible diversity in reproductive anatomy and behaviour. One of the more poorly understood aspects of fish reproduction is the presence, function, and classification of reproductive accessory glands. Akin to the human prostate, these glands vary in form and function, however not all fish taxa have them and their evolutionary history is not well understood. The aim of my research project was to further our evolutionary understanding of these glands by 1) determining the number of times these glands have evolved and 2) exploring the co-evolution of parental care traits, and the presence of alternative reproductive taxa. Data on fish parental care and mating behaviour was collected from peer-reviewed sources across 361 species. Modern comparative phylogenetic methods and discrete character evolution models were used to determine how many times across a phylogeny the traits of accessory glands and parental care evolved. In addition, Pagel’s 1994 test for correlated evolution confirmed correlation between the two traits (p=0.001). The final results of the analyses should provide a better understanding of the evolutionary history of reproductive anatomy and behaviour in this incredibly diverse group of taxa.

Meea Fogal

Consultants: Dr. Josie Libertucci & Dr. Elena Verdu

Investigating the efficacy of anti-proteases produced by food-grade bacteria on attenuating proteolytic activity in vitro

Crohn’s disease (CD) is a chronic inflammatory condition characterized by patchy inflammation and injury along the gastrointestinal tract. With no known cause or cure, current treatments focus on maintenance of remission, often with low efficacy. This drives a need for alternative and more targeted treatment strategies. Upregulation in expression and activity of bacterial and host proteases, which can act to induce intestinal inflammation and damage host tissues, has been demonstrated in patients with CD. This suggests that proteases may contribute to disease pathogenesis. Thus, administration of anti-proteases may help to modulate inflammation. The purpose of this study was to measure the in vitro anti-proteolytic effects of food-grade bacteria expressing anti-proteases. Total proteolytic activity was measured in fecal samples from healthy patients and patients with gastrointestinal inflammation with and without the addition of the supernatant from Bifidobacterium longum, overexpressing the serine anti-protease serpin. Fecal samples were combined with azocasein substrate, PBS buffer and B. longum supernatant, standardized to trypsin, and absorbance was measured at 366 nm. Proteolytic activity with and without B. longum supernatant was compared to proteolytic activity of the assay reagents (control), PBS (blank), and B. longum supernatant. We found that serpin expressing B. longum was able to decrease total proteolytic activity, on average, from approximately 0.70 units of trypsin/mg to 0.49 units of trypsin/mg. These results indicate that bacterial vectors may be a promising route of administration of anti-proteases for the purpose of modulating proteolytic activity. Further investigations are needed to test their anti-proteolytic effects in vivo.

Serena Formenti

Consultant: Dr. Alan Dickin

Mapping Neodymium Model Ages of the Quebecia Terrane near Saguenay, Quebec

The Grenville geological province is located throughout Ontario, Quebec and the eastern United States, covering over a million square kilometers. Since its formation, the area has experienced significant orogenic events; among them the 1.1 Ga Grenville Orogeny. This event rendered older crustal fragments and their associated geological history challenging to delineate by disturbing the original geological features. This study aims to map crustal formation ages within the Quebecia terrane, located along the southeastern part of Quebec. Previous work suggested the presence of slivers of older Paleoproterozoic crust (>1.65 Ga) within younger Pinwarian crust (1.5 Ga). The present study involves the analysis of four geological samples that were obtained from the southern side of the Saguenay graben, where there were thought to be slivers of Paleoproterozoic crust. Analysis through model ages derived from neodymium-samarium radiogenic dating aimed to further delineate the boundaries of slivers of old crust within Quebecia. The presence of Paleoproterozoic slivers within the Quebecia implies that the terrane can be described as a composite arc belt, consisting of accreted oceanic arcs with interaction from older crustal components. Determining the model age distribution within the terrane allows for further delineation of the geological history of the region. The model ages derived from this study fall within the range of Pinwarian aged crust, indicating that slivers of old crust are not present within the study area. These results provide further constraint in the detailed structure of the Quebecia composite arc belt and the geological events preserved within the Grenville Province.

Juliette Froelich

Consultant: Dr. Giuseppe Melacini

Preparing TEV Protease for Protein Purification

Large quantities of protein are produced for industrial usage and research by cloning the recombinant DNA transformed into bacteria. The cloned protein must be separated from the cells’ contents prior to applications and studies. A common purification technique is through binding an affinity tag to the protein; however, this tag must be removed before the protein is considered pure. A viral protease derived from tobacco etch virus (TEV), TEV protease, is used to cleave affinity tags from recombinant proteins. To prepare TEV protease, the TEV plasmid was transformed into E. coli pRare cells using the heat shock method, and the incubated colonies were inoculated in amp-cam-LB media. TEV expression was induced by the addition of isopropyl β-d-1-thiogalactopyranoside (IPTG), an allolactose molecular mimic. The cells were lysed and run through a nickel column by fast protein liquid chromatography (FPLC), with imidazole eluting the TEV protease after filtration. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) confirmed the purity of the TEV protease in fractions 16-22. The concentration, determined by UV280, was 0.683 mg/mL, sufficiently greater than the accepted 0.5 mg/mL. Aliquots of 1 mL TEV protease were flash-frozen and stored until future use in purifying proteins. The activity of the TEV protease can be assessed by binding the protease to a recombinant protein of interest and purifying it through column chromatography. The acceptable concentration and purity of the TEV protease permits the use of the protease in purifying proteins for structural and functional studies.

Elysia Fuller- Thompson

Consultant: Dr. Sean Carey

The Characteristics of Dissolved Organic Matter within Several Streams along the Dempster Highway, Yukon

Twice the amount of carbon currently found in the atmosphere today is stored in Northern permafrost regions. Of this carbon pool, dissolved organic carbon (DOC) is one of the most mobile of carbon pools in the environment and is a major driver for stream metabolism. Measuring DOC quality and quantity is a way of understanding the sources and processes of carbon within the hydrological system.

 In 2018 and 2019, 14 headwater streams along the Dempster Highway that travel through multiple ecoregions, geologies and varying levels of permafrost were sampled. Dissolved organic carbon concentration, stream characteristics, and optical properties were observed and examined for spatial and temporal trends using descriptive statistics. DOC composition was assessed by four fluorescent indices.

 Temporally, DOC followed the well-established trend of peak concentration during spring freshet with similar values from within the literature whereas the fluorescence index generally increased over time, indicating an increase of microbially derived carbon. Fluorescent indices indicated greater microbially derived carbon than previously suggested in the literature, as indicated by a higher Freshness index and Fluorescence Index. Specific Ultraviolet Absorbance remained relatively low, indicating younger DOM which is consistent with measurements from similar environmental regions. Overall, the evidence concluded that Northern headwater stream had more limited seasonal changes in organic matter than shown by other studies, suggesting that the source, age and quality of DOM remains consistent annually within each stream, but that variation was more pronounced between streams.

Jonah Gautreau

Consultant: Dr. George Dragomir

Modeling Maximum Efficiency in Orbital Trajectories

Placing satellites in orbit is an important industry where resource efficiency is a critical issue. This project investigated the topics of space flight into orbit and finding the maximum efficiency with which it can be done. The focus will be on the maximum efficiency with which orbit can be reached. As such, the goal twofold: 1) to determine what the most efficient orbital path is, and 2) how different rocket parameters affect this path. The approach is based on a model for rocket launches that computes a function of rocket position calculated from iterated integrals of acceleration. First, the model determines the best method for calculating efficiency of the rocket for a given rocket shape and thrust configuration, either bell nozzle or aerospike nozzles. The program then determines how different parameters such as thrust profile and drag affect the path taken by the rocket. From this program the expected result is a path that maximizes mass fraction that reaches orbit, and minimizes time spent getting to orbit. The results of this project demonstrate possible gravity turn trajectories, along with the difficulty of finding the true optimal path. There were however limitations on time and computing power in order to compute some aspects.

Emily Heming

Consultants: Dr. Joseph Kim & Dr. Faria Sana

Question Order Effect: What students prefer vs. what is best for them

Taking exams is a crucial part of a student’s academic experience as it is the primary method to evaluate the material learned throughout the semester. However, there is limited empirical evidence on optimal student behaviours when completing an exam. Being able to understand the most effective way for a student to complete an exam can have great benefits for both students and professors. The purpose of the current literature review was to determine two main objectives: (1) In what order do students prefer to answer an exam: do students answer questions based on the order in which they are presented, or based on the difficulty or style of the question? (2) What strategies will have the greatest positive impact on their performance on the exam? The preliminary results of this review suggest that many students tend to complete the easier questions first and not necessarily in the order that the questions are presented. This has, however, not yet been shown to increase the performance of students on the exams. Moreover, students were more optimistic about their performance if they completed the questions in order of increasing difficulty. This bias can potentially have a negative impact as it gives students a false sense of how they have performed. This review can shed light on how students can optimize their performance in order to achieve the best possible grade, and how professors can ensure that the exam is written in a way that can best optimize and evaluate students’ performance and knowledge.

Relevant Sources: (Harvard Anglia Ruskin)

Pettijohn, T. and Sacco, M., 2007. Multiple-choice exam question order influences on student

performance, completion time, and perceptions. Journal of Instructional Psychology, [online] 34(3), p.142+.

Tal, I.R., Akers, K.G. and Hodge, G.K., 2008. Effect of Paper Color and Question Order on

Exam Performance. Teaching of Psychology, 35(1), pp.26–28.

Weinstein, Y. and Roediger, H.L., 2012. The effect of question order on evaluations of testperformance: how does the bias evolve? Memory & Cognition, 40(5), pp.727–735.

Geetha Jeyapragasan

Consultant: Dr. Niko Yiannakoulias

Assessment of the Impact of Lead Exposure on Intelligence in Canadian Children and the Cost of Remediation

A developing concern within environmental health is the impact of lead exposure on cognitive functions, and a number of social and economic impacts. One of the most prominent of those is the effect on intelligence, posing concerns for the future earning potential of the Canadian pediatric population between ages 6 and 11. Numerous studies have used observational study designs and regression modelling to assess the effects of lead exposure on Intelligence Quotient (IQ) scores with varying statistical rigor; however, significant discrepancies exist due to the vast number of confounding variables which affect IQ. The purpose of this project is twofold: (1) understand the independent impact of lead exposure on intelligence, and (2) conduct an economic assessment of the impact of exposure on the Canadian population, measured against the cost of remediation. This was accomplished by conducting a meta-analysis of 12 studies that employed multi-linear regression analysis assessing the relationship between blood lead levels and full scale intelligence quotient (FSIQ) scores. An estimate of the quality-adjusted effect of blood lead level on IQ was then generated, assessing the change in FSIQ scores with every 1 μg/dL increase in blood lead levels. The generated coefficient was then used to assess the total number of FSIQ points lost among Canadian children, and the potential income lost due to this decrease. These values were then compared against the cost of national environmental lead remediation to reduce the entire population’s blood lead level to 0 μg/dL. Using this approach, it was determined whether environmental lead remediation was a cost effective approach to improving intelligence and public health through comparison against other public health, environmental, and social interventions. 

Kate Kim

Consultant: Dr. Karen Choong

Comparison of Enrolment and Data Collection Rates During Pre-Implementation and Post-Implementation Period for the PICU Liber8 Study


In Pediatric Intensive Care Unit (PICU) at McMaster Children’s Hospital, the impact of the rehabilitation bundle, “PICU Liber8”, designed to prevent PICU-acquired complications (PACs), was measured through a mixed methods implementation study. This involved recruiting participants and following up at different time points to measure the impact of the Liber8 bundle during and after discharge from PICU. A key aspect of feasibility in this research is to determine optimal approaches that may enhance patient consent, mechanisms to optimize patient follow-up and processes to ensure the completeness and quality of the data collection.


We hypothesized that consistent and participant-focused approaches would optimize both enrolment and data completion rates during the post-implementation period in comparison to the pre-implementation period. 


The aim of this sub-study of the PICU Liber8 project was to understand how different methods and timing of communication with patients during recruitment and follow-up processes impacted the enrolment and data collection rates.


The working logs recorded in excel files throughout the study were utilized for the enrolment and data completion rates. The methods and frequency of the approaches taken for the recruitment and follow-up processes were compared to support the differences in the outcome.


This sub-study will provide insight for recruitment and follow-up approaches that may enhance enrollment and follow-up compliance of participants, and optimize the quality of data collection.

Frances Lorenz

Consultant: Dr. Paul McNicholas

Machine Learning Approaches to Mammographic Tumour Classification

Breast cancer is the second most common cancer diagnosis worldwide, reaching a total of 2.09 million cases in 2018 (Bray et al., 2018). Regular mammogram screenings (recommended every 1–2 years) for women 40–73 years of age can reduce breast cancer mortality by 40 percent; however, higher testing rates place a greater burden on radiologists (Seely and Alhassan, 2018). The present study applies two advanced machine learning approaches, extreme gradient boosting (XGBoost) and convolutional neural networks (CNNs), to classify mammogram images as ‘Callback’ or ‘No Callback’. The primary goal is to identify benign cases that do not require the attention of a radiologist (No Callback) versus cases which require further investigation (Callback). This will serve to improve clinical efficiency and reduce radiologist workload by streamlining cases of interest. This project uses a curated and standardized version of the Digital Database for Screening Mammography (DDSM) which includes 778 images: 724 ‘Callback’ and 52 ‘No Callback’. The efficacies of the two machine learning approaches are compared and contrasted. Considering the imbalanced dataset, both approaches were limited in their ability to predict the minority class, resulting in low sensitivity. A CNN approach vastly outperforms XGBoost when considering specificity rather than just accuracy. XGBoost achieves 90% accuracy (0% sensitivity, 100% specificity) while a CNN achieves 72% accuracy (31% specificity, 76% sensitivity). Future recommendations include curating larger open source datasets to improve existing approaches and using the two techniques to explore a multi-class problem (i.e., malignant, benign with callback, and benign without callback).

Brittany Mascarenhas

Consultant: Dr. Sigal Balshine

Do males that adopt alternative reproductive tactics differ in aggressive behaviour?

Invasion of non-native species is a pertinent issue, drastically affecting the economy and diverse ecosystems that exist worldwide. One invasive species of international concern is the round goby (Neogobius melanostomus), a small fish that has invaded all of the Great Lakes of Canada and is currently invading Western Europe. Round goby are contaminant and disease vectors in food webs, they outcompete natives, and predate on the offspring of native fish in their invaded regions. While round goby behaviour has been well studied, an area that is not well explored in round goby is their male alternative reproductive tactics. In this study we focused on two specific reproductive tactics, the guarder and sneaker tactic, and explored whether aggression differs between males that display them. We expected that guarder males would display more aggression than sneakers, due to their large size, territorial nature, and different approach to reproduction, which requires physical competition for females. We tested aggression using two assays: a mirror assay and a social assay. Our results revealed no difference in aggressive behaviours between the tactics towards mirrors, but guarder males performed significantly more aggressive acts towards conspecifics than did sneakers. Having a deeper understanding of the behavioural variation of this invasive species will improve our ability to assess its impacts and improve management practices. Thus, the findings of this study and further related future investigations may allow for the development of more efficacious and targeted mitigation strategies in the future.

Aakanx Panchal

Consultants: Dr. Barney A. Geddes, Dr. Turlough M. Finan

Characterizing symbiotic phenotypes of legume-associated rhizobacteria harbouring minimal symbiotic genomes

Replacing industrial nitrogen fixation—a major source of the greenhouse gas nitrous oxide—with biological nitrogen fixation represents one of many ways to mitigate climate change.

Ensifer meliloti is a rhizobacterium that fixes nitrogen in symbiosis with legume plants like alfalfa in organs known as root nodules. Harnessing symbiotic nitrogen fixation (SNF) to enhance or transfer it to other bacteria or plants via synthetic biology approaches necessitates first establishing the minimal subset of genes required for SNF. To date, the 1.35 Mb symbiotic megaplasmid of E. meliloti, pSymA, has been refined to produce “minimal symbiotic genomes” of substantially reduced size (i.e. minSymA1.0, 2.1, 3.1, 3.2).

Legume plant hosts for symbiotic phenotype characterization in this study comprise Medicago sativa (alfalfa), Melilotus albus (white sweet clover), and Melilotus officinalis (yellow sweet clover). Plants were inoculated with E. meliloti strains containing minimal symbiotic genomes. Measuring shoot dry weights allowed quantification of the accumulated effect of SNF ability. Tracking root nodule number over time enabled analysis of nodule formation kinetics.

Since E. meliloti genes required for symbiosis with M. albus and M. officinalis have not been as well defined as for M. sativa, M. albus and M. officinalis may show impaired SNF. If the impairment is drastic, previously uncharacterized genes may surface as genes with important functions. If the impairment is negligible, the minimization will be a step towards the minimal subset of genes required for SNF, a suitable framework for creating synthetic symbioses for a more sustainable alternative to industrial nitrogen fixation.

Matthew Pocrnic

Consultant: Dr. Duncan O’Dell

An Exact Solution to the Bosonic Josephson Junction, and a Review of Analogues of Hawking Radiation in Bose-Einstein Condensates

Here I present a theoretical investigation of the Bosonic Josephson Junction, a many-body quantum system in a double well potential. The system is comprised of a collection of bosons held in traps, and cooled to some very low temperature. Resultingly, the bosons all share the same lowest quantum state, the ground state, thereby forming a Bose-Einstein condensate (BEC). The system functions analogously to the Josephson effect between two coupled superconductors, in which electrons pair up to form cooper pairs that tunnel between the two superconductors. Within the BEC system, we begin with an initial state that describes two condensates in their respective wells. These two BECs, which have not seen each other before (incoherent), are suddenly allowed to quantum tunnel between the two wells by lowering the inter-well barrier slightly. This leads to sloshing dynamics between the two wells, which I compute using Python to exactly diagonalize the Hamiltonian. A series of plots are then constructed to explore the properties of the solutions. Within this investigation, we also include a theoretical discussion on the use of Bose-Einstein condensates as a potential analogue to understand Hawking radiation. The interest in this system is due to its potential to be studied in laboratory.

Angela Pollinzi

Consultant: Dr. Todd Hoare

Synthesis of Magnetic Microgel Particles for Biosensing Applications

This abstract has been omitted from our website due to a confidentiality agreement between the student and the supervisor.

Dana Price

Consultant: Dr. Jonathon Stone

Impacts of hyposalinity on the early development of Strongylocentrotus droebachiensis

As climate change progresses, ocean temperatures continue to rise. These temperature changes have significant impacts on the environment; one such change is the melting of polar ice caps. Polar ice is frozen freshwater. As the ice melts into oceans the freshwater decreases local salinities. This study investigates the impacts of hyposalinity on a cold-water sea urchin species, Strongylocentrotus droebachiensis (green sea urchins). Specifically, the focus of the study is the development of S. droebachiensis zygotes to larval stages and the overall survival of individuals in various salnities. Previous studies have found that this species has the greatest developmental success at 30 Practical Salinity Units (PSU). The tested salinities in the current study include 35 PSU, 30 PSU, 25 PSU, 20 PSU, 15 PSU, and 0 PSU (freshwater) group. Zygotes in these groups were stored at 10 oC in 200 mL of water. Once individuals reached the gastrula stage, they were transferred to beakers containing 200 mL saltwater solution. These beakers were placed in an agitator at 12 oC and the contained individuals were fed with algae. The developing zygotes faired best in 25 and 30 PSU. The further the salinity level was from the survivable salinities, the slower development occurred, and the earlier individuals became damaged and development was arrested. By understanding the impacts of changing environmental conditions on wildlife, we can learn how to protect them from potential dangers like climate change induced hyposalinity.

Caitlin Reintjes

Consultant: Dr. Jurek Kolasa

Investigating Sustainable Design Solutions for Green Urban Living

Over half of the world’s population currently inhabits urban areas. While urbanization is promising for economic productivity, it is also a threat to flora and fauna, puts a strain on energy demands, and can contribute to global temperature increases. To combat these issues, technologies and novel solutions are being implemented in urban areas, with many citizens and governments understanding the need for greener cities. The literature shows that green urban living has various health and environmental benefits and can be used to promote environmental education, recreational activities, and aesthetics. Despite a wealth of evidence indicating that green urban living is beneficial, these initiatives have not yet been sufficiently integrated into modern urban living and planning. This literature review investigates existing green urban living technologies and solutions being tested globally. The objective is to perform a critical appraisal of these technologies to determine promising avenues to pursue for the sustainable development of green urban living within a household. This will include an analysis of the feasibility of various projects by taking into account the needs of citizens, energy demands, and costs of the initiatives. Findings from this study may allow researchers, urban planners, and citizens to work to effectively implement green urban living into their lifestyle.

Maria Romano

Consultant: Dr. Gregory Slater

Toluene Biodegradation in Varying Soil Depths

Toluene is an aromatic hydrocarbon added to gasoline and found in crude oil. In the event of an accidental spill, toluene can be released into the environment. Due to its toxic nature, it is important to understand how toluene will be degraded in the environment. This project set out to determine if there is variability in the degradation of toluene at depths of 0cm, 20cm, and 40cm in a soil profile. Samples were collected using instruments sterilized with methanol and placed into burnt 500mL glass jars. This prevented the introduction of microbes from unwanted locations. Triplicate microcosms were prepared for each depth. Microcosms were prepared with approximately 40g of soil in an amber bottle, and 15ul of toluene was added to each bottle. The control for this experiment was three bottles of autoclaved soils with 15ul of toluene. Three blanks were prepared using empty bottles to monitor the toluene levels in the lab environment. All samples were monitored through the use of a gas chromatography flame ionization detection (GCFID) machine. Sets of duplicate injections with a low relative standard deviation (RSD) were used to determine average peak areas. A normal standard curve of concentrations ranging from 2ppm to 250ppm was prepared to convert average peak areas of the samples to concentration. Results showed that there was variability both between different depths of soil and between bottles with the same depth. Speculations for these observations were made including taking into consideration the heterogeneity of the mixtures. It was concluded that there was variation in the degradation rates of toluene when comparing different soil depths.

Katarina Sacka

Consultant: Dr. Nicholas Kevlahan

Compressed Sensing

Compressed sensing is a method for allowing a signal to be reconstructed exactly, or with small error, from a small number of samples. This paper is a literature review which explains the basic mathematical concepts and assumptions that are behind compressed sensing. Compressed sensing has been developed to exploit sparsity, which is the observation that signals can be represented exactly (or with small error) when expanded in certain bases. While there are many algorithms that can construct a sparse representation of a signal, in this paper we focus on Matching Pursuit. This algorithm has been explained in detail using a simple example in 2-dimensional Euclidean space. It is then further explained how the algorithm would work for a basis such as the Fourier Basis. However, sparsity is only one of several requirements that must be met in order for there to be no loss of accuracy in the recovery process. Incoherence between the basis representing the signal and the basis used to sample the signal, needs to be maximized. Increasing the incoherence can allow for a decreased number of sampling measurements required for no loss of accuracy during the reconstruction of the signal. This is shown in the random incoherence sampling theorem, which explicitly shows how both sparsity and incoherence affect the number of measurements that must be taken in order to recover the original signal with no loss of accuracy. This theorem will be further discussed in this paper.

Pamela Schimmer

Consultant: Dr. Todd Hoare

Investigating the Mechanism of Ultrasound Cork-Shell Release

Ultrasound has been widely used in medicine as both an imaging and rehabilitation tool.  More recently, ultrasound has been investigated as a trigger to initiate drug release from tunable drug delivery vehicles for treatments of diseases such as cancer. A pulsatile release mechanism would be beneficial in further controlling dose, which could be turned on and off non-invasively.

To achieve pulsatile release with ultrasound, microcapsules were fabricated from poly(lactic-co-glycolic acid) (PLGA) shell (to encapsulate a therapeutic payload) into which is embedded silica beads known as “corks”.  In this project, we propose to determine the mechanism of drug release from these corked microcapsules. We hypothesize that the corks pop out from the microcapsule upon ultrasound application, thereby releasing the drug either as a steady-state release or pulsatile release with multiple ultrasound applications. Alternatively, the corks may resonate at different frequencies from the shell, resulting in a pulsatile drug release as the corks move within the shell that is effectively shut off when the ultrasound was turned off.

Microcapsules loaded with bovine serum albumin were formulated using a technique known as coaxial electrospraying. We fluorescently labelled the polymer shell with Rhodamine 123 and the corks with fluorescein. Once formulated, microcapsules were subjected to ultrasound for increasing times and fluorescence was measured. A small quantity of free corks was detected. In a second assay, particles were subjected to ultrasound for two hours and then filtered to identify free corks. A combination of free corks and shell fragments were detected in this experiment, although this could be a product of the vacuum warping the mesh filter.

Future experiments will repeat these tests and make use of the confocal microscope to confirm the initial results and determine the drug release mechanism. This will aid in identifying starting doses for pharmacokinetic analysis and subsequent in vivo experiments.

Mary Anne Schoenhardt

Consultant: Dr. Patricia Chow-Fraser

Developing an approach to assess the efficacy of the 2016 Phragmites treatment program in Big Creek Marsh using Satellite imageryPhragmites australis is an invasive grass that grows in incredibly dense monocultures and spreads rapidly, having been referred to as one of North America’s worst wetland invaders. Phragmites will outcompete native vegetation, cause drying of the wetland and reduce overall biodiversity. Big Creek Marsh, located near Long Point on the northern shore of Lake Erie, is part of one of the last remaining large wetlands in the region, and has recently seen significant growth of Phragmites and the associated ecosystem degradation. Between the years of 2017 to 2019 Big Creek Marsh and the surrounding region was treated with the herbicide glyphosate, followed by rolling, cutting, or burning of the dead stands. Understanding the effectiveness of this treatment is crucial for efficient and effective application of future treatments. This will be done through analysis of summer and winter images from Sentinel-2 satellites. Using the Support Vector Machines classification on ENVI 5.5 (Harris Geospatial) regions of Phragmites will be identified, along with 9 other vegetation classes. The classification will be performed on the summer images, with the winter ones used to create a mask for commonly confused vegetation classes. Following this, a change detection will be performed between consecutive years to identify any changes in the distribution of Phragmites. This will be done across the entire region of the Big Creek Marsh as well as with the application of mask to highlight solely the regions in which treatment was applied, allowing comparison of the treated regions to that of the entire marsh.

Sarah Scott

Consultant: Dr. Jim Lyons

Effects of Sensory Information Compatibility on Anticipatory Judgements

Multisensory integration of all forms, particularly auditory and visual information, is vital to survival of organisms, with basic implications in detecting and acting on predators, prey, and mates. Two theories exist to explain the connection between perception and action: the perception-action dissociation model, and the planning-control model. In this study, visual and auditory stimuli are placed in conflict to contrast the two theories through completion of two tasks. The first task is primarily perceptual, where participants must anticipate when a moving onscreen target reaches a specified location. The second involves perception and action, where participants must anticipate but also intercept the stimulus in the target zone. The visual stimuli changes in terms of velocity, location, and length of time displayed. Participants are fitted with headphones, which pair the visual stimulus to an auditory stimulus that changes in amplitude and location. The visual and auditory stimuli may result in compatible or incompatible response pairings. Stimulus-response compatibility is determined by the degree to which perception is compatible with the required action (e.g. loud left auditory stimulus is compatible with a fast-moving left visual stimulus). Following integration of compatible and incompatible stimuli, the two theories will predict different outcomes. The perception-action dissociation model predicts that incompatible sensory information would have no influence on the resulting action. The planning-control model predicts that incompatible sensory information will significantly influence action. Comparing two prominent theories is vital to understanding human integration of stimuli and its corresponding effects on action.

Jalen Singh

Consultant: Dr. George Dragomir

Monoventricular cardiac model distinguishing between patients with heart failure with preserved ejection fraction and healthy individuals

Heart disease is the second leading cause of death in Canada, and it is estimated that about 600,000 Canadians are currently living with a subset of heart disease and are prone to left ventricular heart failure (HF). A patient’s HF risk is typically associated with a low ejection fraction (EF), defined as the amount of blood pumped out of the left ventricle during each contraction. However, about half of patients with HF have a high or preserved EF, denoted as HFpEF. As a result, EF alone can not provide an accurate quantitative description of the heart with the complex venous and arterial interplay. A volume regulation graph relating end systolic volume (ESV) and end diastolic volume (EDV) may provide a more accurate description. This study proposes a monoventricular cardiac model that can be used to determine the dependence of EF, ESV, and EDV on specific determinants of ventricular function.  A volume regulation graph was created to study the difference between healthy individuals, HFpEF patients, and HF patients with a reduced EF (HFrEF). Multiple other graphs created with this model helped to distinguish between these three groups. This study identifies a measurable difference between HFrEF patients, HFpEF patients and healthy individuals.

Alun Stokes

Consultant: Dr. Ned Nedialkov

Using Convolutional Neural Networks to Solve Jigsaw Puzzles

In the square jigsaw puzzle problem, one must reconstruct a complete image from a set of unordered, square, non-overlapping puzzle pieces. When pieces are non-square, this problem can be solved with O(n2) time complexity – but the specification of square pieces makes this task NP-hard. With no implicit matching criterion in the form of nodules and matching holes on the sides of pieces, this becomes a question of identifying continuity between parts of an image. Here, a fully automated solver is proposed, which makes no use of clues, oracles, or knowledge of the constructed image – which are each often required to adeptly solve puzzles computationally. To do this, a greedy solver is implemented to make placement decisions based on a compatibility criterion given by pair-wise determination by a convolutional neural network (CNN) with gradient boosted random forests (RF). This compatibility criterion is calculated by using a CNN to perform feature extraction on a set of paired pieces, the output of which is then classified using RF trained with gradient boosting. A pairwise accuracy of >99% is achieved on 32×32 pixel pieces. Given the piece-wise solving method, this can be applied to arbitrarily large puzzles, both in number of pieces and pixel size, with the accuracy of the solver determined as a function of the number of pieces.

Megan Tu

Consultant: Nakita Buenbrazo

Constructing a Hyperpermable Strain of E. coli

Gram-negative bacteria are particularly resistant to antibiotics, especially compared to Gram-positive bacteria. Gram-negative bacteria contain a low-permeability outer membrane that restrict the entry of antibiotics into the cell, and efflux pumps that expel any antibiotics that are able to transverse their membranes. Overall, it is the interplay between the outer membrane and the active efflux pumps found within Gram-negative bacteria that characterizes a bacterium’s permeability to antibiotics, as well as other molecules. Escherichia coli is a highly annotated Gram-negative bacterium that poses a significant threat to human health. This project focused on constructing a hyperpermeable strain of E. coli that allows for the free movement of chemicals across its membrane. This project focused on moving a plasmid library created by Alon et al. (2006) into a novel hyperpermeable (HP) E. coli strain.  Each plasmid in the Alon plasmid library contains transformational fusions of a green fluorescent protein (gfp) to each unique promoter in the E. coli K-12 genome, thus the library consists of 1930 unique plasmids.  The plasmids were isolated and transformed into the HP E. coli strain using common molecular biology techniques such as DNA miniprepping and DNA transformation. This project resulted in a collection of E. coli strains that can be used to aid in the discovery of new antimicrobials. This project lays the foundation for The Brown Lab to then probe the novel strain using a variety of chemical compounds, including antibiotics, to determine which promoters in the E. coli genome are activated under certain stimuli.

Yona Tugg

Consultant: Dr. Matthew Miller

Broadly Effective Therapeutics Against Influenza Virus Infections

Influenza viruses are among the most widespread pathogens across the globe. Presently, vaccines are the most effective way to prevent seasonal influenza (flu). Unfortunately, these vaccines provide little to no protection against pandemic viruses. In years where the vaccine composition is a poor match against the circulating strains, its effectiveness decreases substantially. Recent research has focused on developing a “universal flu vaccine” that utilizes broadly-neutralizing antibodies to provide better protection.

We are interested in developing broadly-neutralizing antibodies as therapeutics for those who do not or cannot respond well to vaccination. A candidate antibody was identified using immunoglobulin repertoire analysis of peripheral blood mononuclear cells from vaccinated donors. Preliminary data has shown that the antibody is a broadly-neutralizing antibody due to its ability to bind to the hemagglutinin (HA) of both group 1 and group 2 influenza A viruses. The antibody will be expressed and purified in mammalian cells. Indirect enzyme-linked immunosorbent assays will be used to help characterize the specificity of the antibody by determining which influenza A virus subtypes and HA epitopes that it can recognize. We will test the functional activity of the antibody using microneutralization and antibody-dependent cell cytotoxicity assays to assess its neutralization capability and ability to invoke immune effector cell functions. 

The goal of this project is to develop an antibody-based therapeutic for severe influenza virus infections. It is crucial to understand the mechanisms by which this recombinant broadly-neutralizing antibody confers protection before it can be evaluated in a clinical setting.

Isla Turcke

Consultant: Dr. George Dragomir

Tensile Testing and Analysis of Surgical Knot Types

Mathematical knots are fascinating, geometric structures with numerous properties that allow us to better understand real-world knots. Knot theory is pertinent to any area of our lives in which knots are tied including rock-climbing, sailing, knitting, construction, and suturing.

Suturing is one application where knots are vital to a person’s health. As patients remobilize, sutures experience tension. Depending on the type of knot, tension can cause it to break, slip, tighten, or elongate. If a knot breaks or slips, the sutures can come undone, preventing the tissues from healing properly and potentially leading to infection. Sutures that are too tight increase risk of tissue strangulation and crosshatched scars, while knots that elongate can loosen sutures.

My study explored how different surgical knots behave under an applied tensile force using a tensile machine. It was observed how the knots compared in tensile strength, elongation, and stability. Tensile strength was measured by the maximum force applied before breakage. Elongation is a measure of increase in length due to the applied force and was quantified using image analysis to measure the displacement of markers on the rope. Stability represents a knot’s resistance to slippage and was measured by the maximum force applied before the knot slips and comes undone.

I hope that the results of this study improve our understanding of different surgical knots. Choosing the best knot for a task is critical; knots are often tied in order to ensure someone’s safety and making the wrong choice could prove to be fatal.

Kasia Tywonek

Consultant: Dr. Pablo Serrano

Exploring adherence rates in the “PRophylaxis Extension for VENous Thromboembolism following major abdominal and pelvic surgery for cancer (PREVENT)” study

Cancer patients who undergo major pelvic or abdominal surgery have double the risk of developing blood clots, called venous thromboembolism (VTE), compared to the general patient population. Blood clots can cause limb pain and swelling, as well as limit short- and long-term survival of patients. One treatment to prevent VTE incidence is postoperative administration of low molecular weight heparin (LMWH) injected daily by the patient, for 28 days after surgery. Previous research from Juravinski Hospital and external literature suggests this method may significantly reduce postoperative VTE incidence. Despite this, there are several factors which have been found to affect a patient’s ability to adhere to the treatment, including painful injections or financial concerns. The objective of this study was to understand the difference in treatment adherence rates between gynecological and general abdominal surgical patients. By identifying areas in which adherence is lower, additional resources or patient education can be provided. Patients were screened after surgery and contacted by phone one, three and six months after surgery. Patients were asked about their postoperative medications to confirm they were sent home with LMWH, then asked if they missed any doses. Full adherence is defined as not missing any days of LMWH administration. Bivariant analysis was used to plot the adherence rates of gynecological and abdominal surgery cohorts. This plot shows the rate of adherence as a percentage of patients in each cohort from 2018-2019. This information can be used to improve postoperative patient care and treatment adherence at Juravinski Hospital.

Peiyao Wang

Consultant: Dr. Martin Stampfli

Establishing an in vivo mouse model of cannabis smoke exposure to investigate lung immune responses

With increased legalization in recent years, cannabis has widely gained social acceptance as well as a substantial number of users worldwide. However, the effects of smoking cannabis on respiratory health and the differences of these effects between sexes remain unclear. Cannabis has been shown to have immunomodulatory properties and its smoke’s composition is qualitatively similar to that of tobacco smoke, which has been closely linked with chronic pulmonary obstructive disease (COPD).

This preliminary study sought to establish a murine model of acute cannabis smoke (CS) exposure to investigate the effects of cannabis smoke on lung immune cell populations. Male and female BALB/c mice were exposed to cannabis smoke using a whole-body smoking apparatus previously used in immunological tobacco smoke studies. CS delivery was validated by comparison of plasma tetrahydrocannabinol and carboxyhemoglobin in CS mice to room air (RA) exposed control mice. Furthermore, plasma corticosterone was quantified to ensure stress levels of CS exposed mice were relatively controlled to eliminate stress-induced inflammation as a confounding factor.

Using bronchoalveolar lavage and flow cytometry data, immune cell populations in the lungs of CS and RA exposed mice were statistically analyzed based on treatment and sex. Although immune responses at the studied four-day smoke exposure period were varied, the results suggest that there are sex-dependent effects on immune cell population composition in CS exposed mice. Overall, this study established an in vivo model for future investigations on the immunological consequences of cannabis smoke exposure, such as long-term exposure studies on potential associations with COPD.

Lelia Weiland

Consultant: Dr. Carolyn Eyles

An Investigation of the Mineralogical Composition of the Lockport Dolostone and Rochester Shale exposed along the Niagara Escarpment in Hamilton, Ontario

This project focusses on the mineralogy of the uppermost lithologic units of the Niagara Escarpment in Hamilton. The escarpment bisects Hamilton and causes many issues as ongoing erosion and failure of the escarpment face impacts urban infrastructure. To understand how the escarpment erodes, the properties of the exposed lithologies must be investigated. It is generally understood that higher proportions of calcite in rock will result in increased resistance to weathering and reduced erosion rates. This study aims to compile information on the mineralogy of the three uppermost formations in the Hamilton region, namely the Ancaster, the Gasport, and the Rochester formations. Samples were collected at three different site locations along the escarpment: the Jolley Cut, Dewitt Road, and the Chedoke Radial Trail. At each site, three samples were taken from the top and base of the formation wherever possible. Of the 54 samples collected, 24 were sent to Brock University for thin sectioning.  These thin sections were analyzed under a petrographic microscope, and snapshots with the microscope taken at nine different locations within the sample. Each photograph was analyzed using ZenBlue software, which allows rapid documentation of area counts for each mineral type in the image. These counts were converted into a value identifying the proportion of mineral types in each thin section which was used to statistically analyze inter-formation variability, intra-formation variability, and site variability in mineral content.  These numerical values will enhance understanding of the mineralogical composition and variability of lithologic units and their susceptibility to erosion.

Maggie Wilberforce, (iSci student), Jacob Navarro, Mila Bjelica, Inna Ushcatz, Emily Hauck, Sloane Kowal, Joyce Obeid

Consultant: Dr. Joyce Obeid

Assessing the repeatability, reliability, and reproducibility of natural killer cell assessment in child and adolescent whole blood

Natural killer (NK) cells are the most responsive immune cell to exercise. In children, NK cell counts are known to increase transiently following a bout of acute aerobic exercise. To accurately identify the magnitude of this increase, the error in its measurement needs to be understood. As such, the objectives of this project were to examine the repeatability, reliability, and reproducibility of methods used in studies examining NK cell responses to exercise. Since standardized techniques have been employed, it was hypothesized that the methods would be sufficiently precise to detect any significant differences in NK cells following physical exertion.

Repeatability, reliability, and reproducibility of NK cell assessment was measured in whole blood. Blood samples were collected at rest from children and adolescents participating in an exercise study. Whole blood was stained using fluorochrome-coupled monoclonal antibodies against CD3, CD56, and CD45. Repeatability was assessed by staining a single 100 mL blood sample and quantifying NK cells in duplicate using flow cytometry. Reliability was assessed by splitting a single sample and staining in duplicate (100 mL each stain) before running flow cytometry. Reproducibility was assessed by collecting and staining blood samples (100 mL each) from the same participant under similar conditions approximately one week apart. The quantification of these measures will provide an indication of the sources of error and variability in NK cell counts. These data will inform the interpretation of results from on-going studies examining the effects of exercise on NK cells in the Child Health & Exercise Medicine Program.

Mateas Winter

Consultant: Dr. Altaf Arain

Assessing the relationship between biometric and carbon eddy covariance flux data based measures of productivity over 16 years in three planted forests in southern Ontario.

Planted deciduous forests have become widespread across North America as a pillar in the forestry industry, increasing the need for scientific research to understand their impacts. We hope this study will increase the understanding of carbon sequestration in afforestation and reforestation sites through the assessment of carbon pools at the ecosystem or tree level. The study is extremely valuable due to the extensive overlap of both data sets. The relationship between carbon sequestration from biometric measurements and carbon eddy covariance flux data will be assessed in three planted white pine (Pinus strobus) forests in southern Ontario, Canada. The forests were planted in 2002, 1974, and 1939 and were collected during a 16-year period between 2004 and 2018. The eddy covariance data was collected using flux-tower stations in each forest. This data included the net ecosystem productivity, gross ecosystem productivity, and the ecosystem respiration. Biometric data collected included height and diameter at breast height. Allometric biomass equation constants have been previously developed for these forests by Peichl et al. in 2007. Allometric equations using these constants were used to quantify carbon content from the calculated mean diameter at breast height of each forest each year. Carbon sequestration from the three forests were analyzed and plotted with the eddy covariance data. The long term and interannual trends will be analyzed and the role of these forests in the future will be discussed. We hypothesize there will be some inter-annual coherence between the biometric and eddy covariance-based measurement of ecosystem productivity.

Jonathan Zaslavsky

Consultant: Dr. Jose Moran-Mirabal

Solution-based Benchtop Fabrication of Structured Metal Thin Films

Fabrication of thin films exhibiting micro and nanosized structures is fundamental for developing technologies in the fields of biosensing, energy storage, and materials engineering. The typical approach to producing structured thin films relies on expensive physical or chemical deposition processes, such as sputtering or evaporation, which are often wasteful and require an excess of source material. To address these issues and improve the accessibility and cost-effectiveness, a fully solution-based procedure for the fabrication of structured thin films was investigated. The benchtop technique involves drop casting a polydopamine coating to facilitate the adhesion of nanoparticles and/or the nucleation of metal thin films on a shrinkable polymer substrate. Computer-designed electrode patterns were cut and masked with vinyl, forming the starting point for all subsequent steps. Metallization experiments were carried out using ionic solutions of silver and gold, where the optimal procedure for gold nanoparticle synthesis and metallization was developed, dependent upon the size of the electrode. The surface characteristics of all fabricated metal thin films were assessed using atomic force microscopy, the conductivity was assessed via four-point probe measurements, and the electroactive surface area was determined through reduction and oxidation of the gold surface using cyclic voltammetry in a dilute acid solution. Electrodes were evaluated both before and after shrinking, and compared to sputtered electrodes of the same film thickness, where it was found that the added surface area for shrunken thin films contributed to a greater electrochemical response. The work demonstrated here provides a low-cost, scalable solvent processing technique that can be further developed for thin film coatings consisting of other nanoparticles and metal oxides.