Award Recipients: 2022 Exploration

OBJECTIVE: The primary goal of this multidisciplinary proposal is to develop a platform for growing cultured meat under simulated microgravity conditions.

RESEARCH APPROACH: We will grow cells on specialized microcarriers and expand them in a stirred bioreactor. Microcarriers will be produced in a microfluidic particle generator to achieve size uniformity. A clinostat with an integrated environmental chamber will be developed to generate time-averaged simulated microgravity while maintaining suitable cell culture conditions. We will use induced pluripotent stem cells derived from lamb as starting cells and formulate an animal-component-free media to support the proliferation and differentiation of iPSCs into muscle cells. We will use this platform to systematically evaluate the effect of microgravity on meat cell production parameters including cell expansion, recovery, and quality.

NOVELTY AND SIGNIFICANCE: Human ambition for long-term space exploration and planetary outposts requires ways to provide food for astronauts. Breeding animals in space for meat production is implausible due to the inhabitable environment in space, logistical complications of sending and keeping large animals on other planets, and high cost. Producing meat from cells, a process known as `cultured meat', has already been successfully applied on Earth. This approach allows for producing large amounts of meat from a limited number of cells without the need for significant land use. Further, this system can be integrated into a closed-loop configuration to recycle or regenerate cell media and cells with minimum dependency on resupply from Earth. To achieve this ambition, in 2021, Aleph Farms, which specializes in cultured meat sent meat cells to the International Space Station. Regardless of the outcome of their efforts, the feasibility of performing meaningful research and development activities in space has several practical challenges including high cost and inaccessibility. This proposal addresses these challenges by introducing the first-ever earth-bound tool for producing meat cells in bioreactors and under microgravity conditions. The outcomes of this high-risk/high-reward grant can not only reduce the costs associated with research and development activities related to scalable meat production but also lead to the design of engineered tissues to understand diseases related to low gravity including bone and muscle loss and brain disorders.

Smart, inflatable, and easy-to-make robotic muscles and crawlers, which demonstrate a dramatic shape change with effective cargo transportation and locomotion capabilities, are among the astonishing feats of future technology. Drawing inspiration from the ancient art form of paper folding and the embodied intelligence found in nature, we will introduce origami-inspired sensoriactuator robots with transformable architectures that are magnetomechanically-controlled for multimodal deformation and haptic pressure and temperature sensing. The sensoriactuator robots offer remote actuation, which obviates the need for wired conventional driving units, and enable simultaneous tactile sensing. These traits make the design simple and compact, while impart intrinsic feedback control for precise actuation and adaptive reconfigurability in complex trajectories without relying on vision sensors. The robots will be made of magnetically responsive smart materials, fabricated by adding nano/micro-sized magnetizable hard-magnetic (actuation) and polarizable piezoelectric (sensing) particles into a soft matrix. The correlation between geometrical features and shape morphing characteristics of the soft robots will be studied by adopting reduced-order models and nonlinear finite element simulation. Subsequently, actuation patterns and topological variables of the architected robots to follow a reference trajectory and to provide the desired pressure and temperature sensing functionalities will be iteratively determined by material design prototyping and machine learning algorithms. To meet the underlying challenges in design and manufacturing (e.g. retaining the functionalities of the utilized particles and architectural features, attaining locomotion speeds and accuracies comparable with existing classical hard robots, and corresponding art-science methodological challenge), the proposed research demands the expertise of a diverse team of mechanical engineers, materials scientists, chemists, designers, and computer Scientists. If successful, it will provide the high reward of design and manufacturing of first-of-a-kind active-adaptive sensoriactuator multi-purpose soft robots with integrated sensing features (hence simple and cost-effective) and untethered actuation (using safe, fast, and effective non-contact magnetic control) that can operate in confined spaces with limited energy resources (e.g. in rescue robots, non-invasive medical operations, and drug delivery systems).

Persons in rural, remote, and other low-resource communities have poorer health outcomes than persons in urban centers. Although multiple factors contribute to this discrepancy, inadequate access to medical and molecular diagnostic testing (e.g. clinical labs) is a significant factor. To help address this limitation, we will develop a multimodal smartphone-based platform for cell analysis based on microscopic assessment and molecular markers of health and disease.

A smartphone will be integrated with a small peripheral device that includes a flow chamber for a patient sample (e.g. blood, urine). The device will incorporate multiple cell imaging modalities: bright-field illumination for colour information; dark-field illumination for granularity; and fluorescence for molecular information. Fluorescence imaging of antigenic cell markers will be enabled by designer fluorescent nanoparticles. Each modality will use the smartphone camera for detection. To compensate for the intrinsically limited technical capabilities of the smartphone camera, optical data from each modality will be cross-referenced and deep learning methods will be used to create an efficient and automated cell typing pipeline. Slim convolutional neural networks (CNNs) will perform segmentation, motion tracking, and factor cell size, shape, and optical information (e.g. colour, intensity) into cell classification. All CNN models will be trained using self-supervised learning techniques to increase robustness and pruned to decrease model size. Our main objectives are to develop the device, analysis algorithms, and demonstrate capability for identifying and quantifying specific cell types in blood and urine from their morphological, optical, and fluoro-immunophenotypical features. 

The proposed research is novel and impactful in its development of a portable and low-cost platform for multiple cellular analyses that must normally done by sophisticated microscopy or flow cytometry in a clinical laboratory. Examples of tests and diagnoses that will ultimately be possible on the platform include complete blood cells counts, anemias and other red blood cell disorders, immune cell typing for immune disorders and blood cancers, urinalysis for infections and kidney disorders, detection of circulating tumour cells, and more. Such capability will tremendously improve equity and inclusion in health care by making cellular/molecular diagnostic technologies accessible in many low-resource settings.

Glioblastoma, the most common primary CNS malignancy in adults, has no cure and has had no major treatment breakthroughs in the last 15 years. Glioblastoma evades immune surveillance by several mechanisms including downregulation of adhesion molecules on tumour endothelial cells, ineffective presentation of neoantigens on tumour cell membranes, a lack of signaling to activate naïve T cells, and a physiologically hostile tumour milieu. Cellular immunotherapies using chimeric antigen receptor (CAR) T or Natural Killer (NK) cells come primed and activated with specific neoantigens, but in solid tumours such as glioblastoma, they must still overcome the hurdle of homing to the tumour and exiting the vasculature. The goal of this research program is to activate brain endothelial cells with ultrasound and microbubbles (ultrasound contrast agents) in favour of CAR cell therapy. It is hypothesized that ultrasound and microbubbles can be used to enhance the recruitment, adhesion, and extravasation of CAR T/NK cells through the tumour endothelium. We propose a novel in vitro approach whereby human microvascular endothelial cells are cultured under flow conditions with the introduction of human glioblastoma supernatant extracted from surgical patients, with complete molecular and methylation analysis of the tumour. These human tumour endothelial cells will be subjected to microbubbles under the influence of ultrasound, and endothelial form and function characterized using both optical and fluorescent microscopy. Immunohistochemistry will be used to quantify changes in cell adhesion molecules (selectins, VCAM, ICAM), and will be correlated to bubble acoustic emissions. CAR cell binding after ultrasound exposure will be assessed by introduction of human CAR NK-92 cells into the flow chambers, where adhesion can be monitored in real-time. This project combines two cutting edge technologies: therapeutic ultrasound and cellular immunotherapy. The experimental setup is highly innovative, using entirely human cells lines under physiological flow conditions, paired with human brain tumour specimens directly from the operating room, and a custom-designed acoustically-coupled microscope. It will allow direct observation of cellular interactions at the cerebrovascular endothelial layer and tumour endothelium in a way that would otherwise be almost impossible. This will pave the way for new treatment paradigms combining cellular immunotherapy and therapeutic ultrasound.

Objective: Develop a biomimetic patient-specific graft prototype for the treatment of aortic aneurysm.

Approach: a disaggregated database based on ex vivo mechanical and microscopy tests on human descending thoracic aortas will be built, including smooth muscle activation. Swine samples will be decellularized as validated by DNA assays. Induced pluripotent cells from human adipose stem cells will be implanted and differentiated in fibroblasts, smooth muscle and endothelium. The prototype will be characterized by mechanical and microstructure tests. The feasibility of self-donation of stem cells, as well as the acceptance of the graft by followers of religious precepts prohibiting the use of swine tissue, will be evaluated assessing psychologically patients suffering from aortic aneurysm.

Novelty: The graft uses the patient's stem cells, implanted on a swine aorta, never considered before for aortic replacement. It matches the characteristics of the individual, who plays an active role with psychological support. Mechanical properties, diameter and thickness are adjusted to the patient.

Significance: Cardiovascular pathologies are the first cause of death worldwide. Aortic aneurism may affect disproportionally indigenous groups in Canada. Surgical repair after aneurysm replaces the damaged portion of the aorta with stiff Dacron grafts. The difference in mechanical properties causes complications (false aneurysm, heart hypertrophy). The proposed graft maintains the cyclic expansion of the aorta improving perfusion. The decellularized swine aorta has three layers with properly oriented collagen and elastin fibers. The new graft is low-cost, bio-compatible,with no risk of rejection. It eliminates problems caused by present grafts. The patient's stress, risk perception and informational needs will be assessed during graft development.

High risk/reward: While the use of decellularized matrices as scaffolds for tissue engineering has been proposed, it has not been used for aorta replacements. The proposed regenerative approach needs collaboration between biomedical engineers, clinicians and psychologists. It challenges a technology that represented the status quo for forty years. Our holistic approach takes diversity into account in the composition of the research team and in the idea of a bespoke treatment, with attention to the case of indigenous populations and of religious prohibitions against animal tissue.

Intracerebral electrodes are an essential and widely used tool to record from and/or stimulate specific brain areas, in animal research and clinical settings. Current electrode design is limited to orthogonal (linear) probes. Yet, the topography of some brain areas requires development of electrodes with `unorthodox' shapes and orientations. The goal of this project is to develop curved electrodes for high density recording and stimulation of difficultly accessible brain areas.

In mice, the ventral hippocampus (vHP) is located deep in the brain and plays a fundamental role in emotion and memory. The vHP is organized in layers that follow the shape of the skull, ending at a 45° angle relative to horizontal plane. Analysis of vHP electrophysiological properties (such as current source density mapping) requires multi-site recordings perpendicular to vHP layers. Our two-year aims are to develop a prototype of curved electrodes, build an implantation system and test them in the mouse vHP.

Our first aim is to develop a prototype of curved electrode using photolithography, a technique allowing high resolution fabrication of small objects. Conducting polymer coating, processed by electrodeposition or spin coating, will be used to lower impedance. To allow insertion, electrodes will be rigidified using a bioresorbable silk fibroin backbone.

Our second aim is to build and test an implantation device. While linear electrode insertion requires translational movement, insertion of a curved electrode requires rotation. Our implantation system will i) be attachable to a stereotaxic apparatus for precise insertion in the brain and ii) allow rotation of the electrode. Insertion and electrical properties will be tested in vitro.

Our third aim is to validate implantation, localization and recording properties of curved electrodes in vivo. Electrodes prototypes will be implanted in the mouse vHP and localization will be validated anatomically. Curved electrodes will be used to record electrophysiological activity in the vHP.

Our team, composed of a neuroscientist and a material scientist with expertise in electrode fabrication, will work in collaboration with a neurologist and a neurosurgeon with the long-term objective of adapting curved electrodes to clinical applications. In the future, development of innovative electrode shapes and configurations will allow for safer, less invasive treatment of patients requiring intracranial implantations for medical purposes.

Buildings generate nearly 50% of global Carbon Dioxide (CO2) emissions. Hence meeting the United Nations' decarbonization target of zero emissions from buildings by 2050 is an urgent necessity. However, such decarbonization critically requires determining these buildings' energy flow accurately, a process fraught with many practical challenges.

To determine energy flow in a building, physics-based Building Energy Modeling (BEM) software would require an accurate description of its structural form (e.g. geometric dimensions or ratios) and construction materials (e.g. concrete or glass). Such descriptions are obtained typically from building records that may not be available or on-site audits by experts that are very time-consuming and expensive. On-site audits have other limitations. Audit experts usually vary in their expertise, diligence, and visual acuity. Also, inaccessible areas of the building increase the difficulty of such audits.

Here, we will combine our expertise in Urban Sustainability Audits, BEM and Image Processing to develop a paradigm-shifting alternative. We propose using autonomous Unmanned Aerial Vehicles (UAVs) equipped with visible, thermal, and hyperspectral imaging systems. UAVs would obtain a building's geometric properties, material representations and thermal profiles, under different weather conditions. These data would then be used to estimate its aggregated energy flow accurately.

This project will involve three tasks:

1) Purchase and equip an UAV with multi-modality imaging systems: visible, thermal, hyperspectral, & weather sensors.

2) Develop a flyover protocol to collect building data from different urban locations.

3) Obtain the geometric properties (visible imaging), material representations (hyperspectral imaging), & thermal profiles (thermal imaging) of aerially audited buildings from the multi-modality images collected by the UAV. These building parameters would be required inputs to the BEM software package to obtain an accurate estimate of its energy flow.

The highest risk in our proposed research is its significant novelty, where multi-modality imaging-based aerial building audits have not been tried before. However, it could have a very high impact by enabling the achievement of the UN's decarbonization target of zero emissions from buildings in the near future. It further has the potential to engage with various communities to support sustainability measures for a wide range of building typologies.

Around 86,000 Canadians live with spinal cord injury (SCI), with 4,500 new cases yearly. The annual economic burden of SCI on Canada is more than $3.6B. Restoring stable walking function is the highest priority for those with SCI. Although assistive technologies such as wearable exoskeletons and functional electrical stimulation (FES) can help, they are neither widely implemented in Canadian clinics, as their control programs are not adaptive to different SCI conditions (in case of exoskeletons), nor can be used for high dose exercises (in case of FES, due to muscle fatigue). These technologies are typically designed based on a few users' data. Due to the large heterogeneity of the SCI-induced neuromuscular impairments, a technology suitable for one user may be inefficient or have adverse effects on others.

Innovation: Optimizing the neurorehabilitation and prescribed technology for individuals with SCI has been a challenge for neural engineering due to heterogeneity of the impairments. To address this challenge, we will develop a clinically usable method to identify mathematical models of impaired neural control of movements after spinal cord injury and use them in developing a `personalized' neuro-robotic rehabilitation technology to achieve optimal recovery. This will be achieved by bringing together expertise from biomechanics, machine learning, robotics, neurorehabilitation, imaging, and medicine through a multi-institute collaboration.

Objective: To develop a personalized, wearable, assistive-rehabilitative technology to regain stable locomotion for individuals with SCI.

Research Approach: 1) We will characterize a musculoskeletal model for individuals with SCI based on biomechanical measurements. 2) We will design innovative experiments involving mechanical and sensory perturbations applied during assisted walking to identify the affected neuromechanics using system identification and machine learning techniques. 3) We will design the personalized control systems for the exoskeleton as well as functional electrical stimulation based on the neuromechanical models and optimal control theory. 4) We will evaluate the efficacy of this technology for individuals with SCI, by comparing with the existing robotic systems.

Impact: This research will develop new paradigms of personalized neuro-robotic rehabilitation, significantly impact the quality of life of Canadians with a disability and reduce the SCI-related costs.

Bankruptcy is among the most effective policies to ease financial distress. Yet Canadian and international organizations including Ontario's Office of the Superintendent of Bankruptcy (OSB), the International Monetary Fund, and the World Bank have noted barriers to bankruptcy access, including financial costs, technology, and stigma.          

We are working with the OSB to pioneer an online tool to reduce financial, practical, and psychological barriers to bankruptcy for low-income Canadians. Building on this tool, our interdisciplinary team in law, economics, and cognitive science will pursue high-risk/reward research that will generate five papers about low-income bankruptcy access.

Two economics papers will conduct the first randomized control trial (RCT) on lowering the cost of bankruptcy. Canadian debtors must pay $1800-$6500 to become insolvent. We will randomly reduce fees by up to 84% for low-income debtors. Our first paper will use this random variation to estimate factors that affect the decision to file, including fees and the convenience of an online tool. Our second paper will again use this random variation to estimate whether insolvency helps or hinders employment status, spending, and credit access.

Our cognitive science paper will estimate how insolvency affects wellbeing. We will track insolvent vs solvent debtors' wellbeing: specifically, overall life satisfaction, sub-clinical anxiety and depression, and moment-to-moment happiness. Using these diverse methods, we will test whether insolvency lowers life satisfaction due to stigma but provides moment-to-moment relief.

Our final two papers concern Canadian and comparative bankruptcy law. Low-cost bankruptcy is a policy priority for Ontario's OSB that is unrealized due to political, market, and technological barriers. Our Canadian law paper will generate empirically-tested proposals to expand bankruptcy access. Our comparative law paper will examine Canada's use of insolvency trustees instead of lawyers, which the US may follow after 2022's Upsolve v James. The Canadian experience suggests a dilemma: tight trustee regulation can be anti-competitive but loose regulation is vulnerable to predatory actors (Pearce v 4 Pillars Consulting Group).

Our project is high-risk: we require new technology, buy-in from stakeholders, a major RCT, and innovative measures of wellbeing. But the reward is equally high: we can pioneer a new, more humane bankruptcy system in Canada.

Healthy pregnancies depend on proper placental development, an organ formed exclusively during pregnancy to allow blood supply to the growing baby and elimination of fetal waste. Placental formation involves processes of uterine vessels remodeling and angiogenesis, which is the formation of new micro-vessels. They are orchestrated by endothelial cells from both maternal and fetal origin and trophoblast cells through direct (cell-cell) interactions or indirectly through signaling molecules that together drive vessels networking connecting maternal and fetal circulations. However, the understanding of mechanisms disrupting placental vascularization is still very limited, particularly during the early stages of pregnancy when impaired blood flow can lead to the onset of maternal circulatory stress, high blood pressure and fetal growth restriction, characterizing a severe hypertensive disorder called pre-eclampsia (PE). This is relevant because PE reaches one in ten pregnancies, causing the deaths of over 70,000 mothers and 500,000 babies annually worldwide, and for which no reliable option of treatment other than delivery exists.

Advances in this field have been drastically limited by unsuccessful models of early human placental development justified by ethical restrictions and the lack of relevant biological samples capable of clinically reproducing the origin of placental dysfunction. To fill this gap, our main objective is to develop and validate a model of placental vascular unit by combining endothelial and trophoblast cells. To achieve this goal, our group merged experience in the primary culture of human placental endothelial and trophoblast cells with computational imaging to objectively measure the effects of clinically relevant PE-like stressful factors and therapies on cell-cell interactions, patterns of cell migration, cell death and angiogenic responses. This project is unique and has a high risk because it proposes the primary culture of human cells extracted from the same placenta to reproduce in vitro the intrauterine environment, requiring parallel and well-coordinated cell cultures and modeling. Rewards from this project include the delivery of the first human-derived placental vascular unit and the investigation of mechanisms and patterns of angiogenesis during placental development and pathology. Collectively, we expect to accelerate the discovery and testing of new therapies to treat PE and prevent comorbidities in mothers and children.

Background:

Chronic liver disease (CLD) affects an estimated 25% of Canadians. CLD can progress to cirrhosis, resulting in significant morbidity and mortality. Cirrhosis and ongoing liver inflammation impair drug metabolism, making it difficult to prescribe the correct medication dose without side effects. Liver transplantation (LT) represents a cure for CLD. However, fibrosis often redevelops over time, leading to cirrhosis in some recipients. Therefore, LT is a model that can help understand the natural history of CLD in an individual. Digital twinning with Machine Learning Algorithms (MLAs) has been previously used in other clinical settings to simulate how a patient will respond to a given exposure. Liver function is difficult to assess, and could be interrogated by an intrinsically imageable high-density lipoprotein-like porphyrin nanoparticle (PorphyHDL) selective to liver cells.

In order to guide safer therapy in CLD, we will:

1) Develop a MLA incorporating longitudinal laboratory, radiomic, and clinical data to predict changes in functional liver mass (% of total capacity);

2) Examine functional liver mass in a subset of patients using PorphyHDL for molecular imaging;

3) Predict the safety of a medication and dose for a specific individual with CLD using MLAs.

Research Approach:

Our team developed a MLA on longitudinal data from multiple clinic visits to predict death due to liver failure in LT recipients. We will now use a distinct approach with recurrent and convolutional neural networks trained on longitudinal clinical data, along with imaging from LT recipients to determine functional liver mass. This will be measurable by changes in blood levels and dosing of antirejection drugs, and validated in a subset of patients using PorphyHDL as radiotracer. We anticipate that uptake of PorphyHDL into liver cells will vary depending on degree of scarring (which interferes with blood circulation), thereby reflecting functional liver mass.?

Novelty & Expected Significance:

Our approach employing MLAs in combination with nanoimaging is high-risk, but will be high-reward in solving a common problem in clinical practice. Dynamic time-varying data from our closely-followed LT patient cohort will be used to predict liver function changes over time, and validated using an imageable nanoparticle. Digital Twinning of the Liver has the potential to transform care of patients with CLD, by guiding medication dosing to keep them safe from side effects.

Resource extraction activities in the boreal forest occur on Indigenous territories disproportionately affecting food, medicine, and traditional practices of Indigenous peoples. Forestry is one such extractive industry, which employs the large-scale application of herbicide in reforestation. Herbicide application increases the chances of successful monocrop regrowth by eliminating native broadleaf plants, targeting shrubby heathers (Ericaceae), which includes traditional food and medicinal species important to Indigenous communities. Herbicide disrupts a healthy functioning ecosystem, including soil, lingers in the environment, bioaccumulates in surviving or regrown plants, and moves through the food web, affecting animal and human health.

One of these target species is maskêkopak, or Labrador tea (Rhododendron groenlandicum), a medicinal and edible Northern Bush Cree cultural keystone species. The harvesting location, time, and preparation of Labrador tea is embedded in traditional knowledge passed down for generations. Bigstone Cree Nation members are experiencing ecological anxiety about food safety from observing ecosystem damage from glyphosate spraying and have requested to partner with our research team to address their concerns. Thus, we propose research based on Indigenous plant and ecosystem knowledge of Bigstone Cree Nation members to inform microbial sciences. First, a team of Bigstone Cree Nation environmental monitors, youth, and researchers will learn from Elders and Knowledge Holders about Laborador tea locations, use, and traditional practices, including stories and relationality. Guided by this knowledge, the team will sample and analyze the microbiome of the soil and the Labrador tea rhizosphere using DNA and RNA sequencing techniques. The analyses will compare the healthy, unsprayed soil and rhizosphere with that of sprayed locations along a spatio-temporal scale. This will help to establish a microbial measure of ecosystem damage and glyphosate persistence. This novel interdisciplinary approach aims to support the continued use and transfer of Indigenous plant knowledge to generate complementary microbial insights. Microbial markers of glyphosate use and persistence will contribute key evidence supporting Indigenous food and medicine sovereignty, advocating for regulatory changes for herbicide use in forests, and improving the physiological, emotional, and spiritual wellbeing of Indigenous communities across the boreal forest.

Neuroprostheses are electrical interfaces with the nervous system replacing or helping recover lost function after injury or disease. They are applied to reanimate neural circuits and restore motor function by fostering ongoing neural plasticity mechanisms. Movement is a complex computation performed by distributed networks. Yet, today, neuroprosthetic devices are simple and localized: they target single precise brain, spinal or nerve areas, with promising impact that would greatly benefit from synergetic recruitment.

We here propose to develop the next generation of distributed, intelligent systems. The future of neuromodulation medicine is precise multi-pronged interventions with high-density neural interfaces, regulating multiple neuronal networks concurrently at large-bandwidth. This development requires an algorithmic framework to handle optimal treatment delivery. Our solution is autonomous neuroprosthetic AI optimization, controlling the complex dynamics of neurostimulation.

We aim at introducing distributed neuromodulation as treatment standard and large-scale automation as a structural part of neurotechnology design. Combined neurostimulation intervention can take the established rule for neural repair "fire together, wire together" to the highest extent.

In animal models we will demonstrate the power of this approach by solving two unmet neuroprosthetic problems:

1) Network shaping. Paired stimulation of connected neuronal networks can produce powerful potentiation of neural signal transmission, but stimulation control is still an issue. With real-time intelligent control we will unlock distributed neurostimulation protocols resulting in reliable long-term potentiation of motor circuits.

2) Functional motor improvement. Brain and spinal stimulation are complementary for motor control. With distributed neurostimulation we will reverse motor deficits of hand/arm paralysis beyond what is possible with localized neuromodulation.

The key challenge of distributed interfacing is complexity. Combining multiple sources of neurostimulation is exponentially harder (the "curse of dimensionality"). Our solution is the theoretical algorithmic novelty of a fully hierarchical optimization framework, capable of combining and linearly separating multiple stimulation sources.

Translation is a core value of our research. From the rat to the non-human primate, our framework is built to scale proof-of-concept demonstrations towards humans.

Vector-borne diseases (VBDs) account for nearly one-fifth of all infectious diseases worldwide, and more than one billion people are infected every year. Climate change will likely continue to drive VBDs range expansion, increase the duration of transmission seasons and lead to VBDs-related epidemics. Most control measures involve the use of widespread insecticides or individually applied repellents that interfere with molecular level olfactory and/or gustatory processes of the insect vector. Evolving resistance and unintended adverse ecological consequences have prompted an urgent need for new solutions. In this project, we are AIMING to develop an unconventional insecticide/repellent-free approach to prevent VBDs by integrating the interdisciplinary areas of entomology and nanotechnology. We will develop an orthogonal measure for bite prevention and corresponding control of disease transmission by utilizing a skin-safe particle-laden barrier coating applied directly on the skin to prevent insect bites.

We will explore the central hypothesis that inorganic particles with a plate-like geometry can create a physical barrier capable of preventing insects such as mosquitoes from biting a host, thereby stemming the spread of infections by nanomechanics investigations and field research. The use of particle barriers allows for the avoidance of other chemical repellents that pose health risks due to allergenic and other potential side effects and environmental pollution. The prospective IMPACT of the proposed collaboration will entail a transformative effect on mitigating the transmission of VBDs and a broad-ranging impact on public health across the globe. Specifically, the mosquito vector and human host will be selected as the initial model system, though at later stages the proposed concept could be extended to numerous other biting insect vectors and hosts. This could include biting insects such as fleas, sandflies, or ticks that spread various pathogens to humans, livestock, and domestic pets. The same framework could also be applied to investigating sharpshooter insect vectors of plant diseases, which cause hundreds of millions of dollars in agricultural losses annually on multiple continents.

With a surge of renewable energy sources, grid-scale stationary battery systems play a critical role in stabilizing the electricity grid. An efficient battery management strategy involves a challenging optimal control problem that needs to exploit high-fidelity nonlinear partial differential equation models to maximize the economic incentives, mitigate degradation effects, and satisfy critical safety constraints. This complicated control problem cannot be solved online at the frequency required (e.g., seconds).

We have assembled a team of researchers with diverse interdisciplinary expertise in the areas of machine learning, control, optimization, power systems, and electrochemistry to address this problem. We propose to train a decision tree (DT) model offline to approximate the optimal control laws, so that the online computational cost is negligible (e.g., microseconds). The key advantage of DT models lies in interpretability, as such models closely resemble human reasoning and are easy to comprehend by human decision-makers. The operator of a stationary battery is more likely to understand and trust a DT model over black-box models (e.g., neural network).

A critical challenge is the training of optimal DTs. The traditional greedy-based heuristics can only produce sub-optimal DTs. Mixed-integer optimization (MIO) can be utilized to learn provably optimal DTs, yielding significant improvements in testing accuracy over heuristics. However, existing MIO-based approaches can only address a small dataset (e.g., hundreds of samples). We propose to develop scalable deterministic global algorithms capable of learning optimal DTs from large datasets (e.g., a million samples). Our work is based on the observation that DT learning can be reformulated as a two-stage stochastic programming problem, and thus various approaches proposed in the stochastic programming community can be utilized to exploit the problem structure.

This is a high-risk project. We seek to challenge the prevailing mindset that global optimization is not practically relevant to large-scale machine learning problems. However, despite the risk, the project is extremely high-reward. Our algorithm will dramatically expand the application scope of optimal and interpretable models. The advances in battery management will lead to a more sustainable and reliable grid system and support Canada in reaching its international emission reduction commitments.

While there is a well-known gap between what the general public and policy makers understand about science and what is known by experts, this gap is particularly perilous in regards to climate change. Climate change is increasingly recognized as a paramount threat to life on the planet. The most recent report from the Intergovernmental Panel on Climate Change highlights the extreme and worsening impacts of climate change, including rising sea levels, heatwaves, drought, flooding, regional food and water shortages, coastal storm damage, and more. Canada is particularly vulnerable to enormous disruptions: as Canada's Changing Climate Report states, "Both past and future warming in Canada is, on average, about double the magnitude of global warming". Scientists are generating massive amounts of data about climate change and developing significant understandings of the causal factors, wide-ranging projected impacts, and necessary mitigation and adaptation strategies. To know how to respond and make changes both policymakers and the general public need to be better supported to develop actionable comprehension. Currently scientists inform each other via expert publications and conferences. We, as part of the public and policy makers, receive our information via the media and the web - and in our current catastrophic blending of information with misinformation, we are at risk of well-intentionally taking ineffective or even harmful actions and decisions. We need the best and most current scientific information in an easily accessible format that includes data transparency and is also both scientifically informed and verified. To close this gap, we have assembled a team that includes experts in data visualization, narrative construction, data comics, and climate change. We will work collaboratively to develop climate change data comics that combine compelling narratives with comprehensible data visuals that are informed and verified by the appropriate scientists. 

Epilepsy, defined by spontaneous recurrent seizures, is a life-altering disease that currently affects up to 1 million people worldwide. Although many anti-epileptic drugs have been introduced in the last decades, 30% of patients taking anti-epileptic drugs continue to suffer from frequent seizures. These recurrent seizures, along with associated risks of injury and premature death as well as comorbidities such as depression and memory loss, highlight the urgent need to develop new treatments for epilepsy patients.

Typically, pharmacologic targets and treatments for epilepsy are developed and/or tested using non-human models of the brain, as living human brain tissue is exceptionally challenging to obtain and use experimentally. Here, in a specialized collaboration between neurosurgeons, neuropathologists, fundamental neuroscientists, and mathematicians, we will use rare living human brain samples to identify next-generation epilepsy targets and treatments.

In our proposed research, we will identify disease-associated cell populations and their functional role in the epileptic neuronal network. We will test the hypothesis that specific neuronal and non-neuronal cell populations causally mediate the genesis and progression of epilepsy. Using cutting-edge high-throughput biological techniques, we will identify marker genes and proteins to inform our fundamental understanding of epilepsy and help guide clinical diagnosis for epilepsy patients. In tandem, we will use a designer viral approach in human brain slice cultures to target and manipulate specific cell populations. We will use these techniques to understand the causal molecular-cellular operation of the living human brain, detect novel biomarkers for epilepsy progression, and identify novel targets for therapeutic intervention in epilepsy.

Our project, using cutting-edge techniques applied to living human brain samples, is uniquely positioned at the interface of fundamental neuroscience and clinical application. As many elements of this research are unique either within Canada or across the world, this research has the possibility for ushering in an entirely new approach for understanding the fundamental operation of the human brain, as well as treating epilepsy. 

Breakdown and Repair in Gaza's Health Care Sector

Political instability, military conflicts, and blockades plague Gaza.  In its health care sector, the tumult manifests in deteriorating and broken medical equipment unable to be fixed due to a lack of replacement parts, a dearth of skilled technicians, and impossibly restrictive maintenance and repair contracts imposed by far-away manufacturers. 

Objectives

This project seeks to build reparative capacity in Gaza's health care sector by: 1) enhancing understandings of the sociotechnical dimensions of breakdown and repair in the provision of equitable care; 2) studying and documenting the condition of existing medical devices and the needs of the community, and creating a process for prioritizing the repairs; 3) producing open source 3D printed replacement parts for equipment and publishing open access schematics and training materials for use in Gaza and elsewhere; and 4) training local biomedical engineers to do repair work.

Research approach

The challenges and solutions facing Gaza's health care sector are not purely technical, political, economic, or social. Our work is, therefore, guided by a sociotechnical perspective that accounts for the entanglements between these worlds, and girded by extensive practice in emergency medicine and medical device engineering, qualitative methods, design thinking, and legal and policy analysis and advocacy.  Our disciplinary distinctions meld together under a set of shared core principles including a recognition that medical care is a human right, technological and informational tools should be sources of liberation rather than oppression, and sociotechnical transformation is possible only through participatory solution-building.

Novelty and significance

The risks attendant to this project are significant:  we aim to carve new channels of medical equipment repair in an imperiled and resource-strapped region by developing tools and training to empower the community to fix things themselves.  The potential risks are justified, however, by the rewards.  This project will serve fourteen public hospitals, over fifty primary health clinics and the two million patients they serve. Every repaired medical device will save lives and improve health outcomes.

The crisis in Gaza is unique, but breakdown is not.  The ability and right to repair things is critical for environmental sustainability, economic viability, and the welfare of societies around the globe.   

Urinary tract infections (UTI) are one of the most common infections in humans, affecting women disproportionately. The current paradigm on its pathophysiology is the ascending route of uropathogens from the perineal area into the urine bladder. Recurrence of urinary tract infections is believed to occur because of urothelial intracellular bacteria grouped in biofilm-like structures that free-float in urine once urothelial cells exfoliate.

Our preliminary data shows that between 85-95% of urothelial cells in urine contain intracellular bacteria communities (IBCs) compatible with biofilm-like structures. Importantly, these IBCs are present in participants with recurrent UTI and in healthy controls. We can also identify free-floating IBCs using scanning electron microscopy.

We hypothesize that UTI are secondary to exfoliation urothelial cells containing IBCs. The occurrence of UTI depends on inability of the urine bacteriophages to inhibit uropathogens dominance

With the support of New Frontiers in Research Fund we plan to confirm this hypothesis and perform the following studies:

1) Prospective and cross-sectional studies: We will confirm our hypothesis including participants in the general population (prospective case-control cohort, being cases participants with recurrent UTI and controls without) and participants with kidney transplant (cross-sectional case control-study and prospective cohort). We will use ImageStream to quantify urothelial cells containing IBC from urine samples. In kidney transplant participants we will identify the source of urothelial cells in sex-mismatch kidney transplants staining the Y chromosome.

2) Microbiological studies: We will isolate urothelial cells from urine and sequence intracellular bacteria. This substudy will be done in participants in prospective cohorts and genomic analysis of intracellular bacteria will be compared with new episodes of UTI.

3) Microbiome studies: We will study the composition of the urine microbiome in participants in prospective and cross-sectional studies, including the urine virome, by deep sequencing metagenomic. We will prove that the population of urine bacteriophages is decreased in participants with recurrent urinary infections.

The proposed team has experience with ImageStream, metagenomic and cell imaging, which ensures success of the project.

This study will change the current paradigm of UTI and open innovative lines to improve current management strategies.

Primary bone tumors of the spine (PBTS) are rare and challenging tumors. For most aggressive and malignant bone tumors, an en bloc resection is recommended.  Appropriate surgical margins in the spine are obtained roughly in 75% of time. Positive margin may alter the prognosis and treatment. The interpretation of the margins is complex and subject to sampling error. It would be ideal to have a reliable alternative to determine if the tumor has been completely resected.

The genomic landscape of PBTS is heterogenous and require tailored biomarkers. Using whole exome sequencing to identify somatic mutations that then become the basis for a personalised biomarker, it is now possible to detect circulating tumor cell DNA (ctDNA) in the pre-operative blood in over 87% of chordoma patients. These novel technologies have the potential to revolutionize treatment and surveillance algorithms by introducing a reliable personalised biomarker to guide treatment and to detect tumor recurrence in a timely fashion.  A potentially clinically useful application of this technology is to detect microscopic residual disease after surgery. Whether it can be applied to other primary bone tumors of the spine or detect residual disease after surgery needs to be determined.

Up to 10% of the primary bone tumors are initially misdiagnosed and re-classified after thorough molecular testing. This highlights how difficult these tumors can be to label and the limitations of the classical histopathology in the modern era of medicine. Misdiagnosis can have tremendous impact. Due to the rarity and complex genomic landscape, there is no off the shelf mutational panel that can assist with the diagnosis.

Through a collaborative effort, we propose to investigate personalised biomarkers for PBTS and their ability to detect residual disease after surgical resection. This will be done using whole exome sequencing and personalised PCR assay on blood sample collected pre-operatively and post-operatively. Being major referral centers and with comprehensive biobanking protocols, we are in a unique position to study theses tumors. This would be the first step in establishing novel biomarkers. Lastly, we proposed to develop and validate with our PBTS cohort (including benign and malignant tumors), a primary bone tumor specific molecular mutation panel to assist and improve the diagnosis accuracy to ultimately improve patient care.

OBJECTIVES: This interdisciplinary project bridging Health and Engineering will develop a novel way to implement the ultra-high dose rate (UHDR) radiotherapy (RT), often coined as `Flash'. This approach can drastically shorten the treatment times (from ~25 days of hospital visits to a single trip), and markedly improve treatment outcomes. Preclinical experiments showed that Flash provided excellent normal tissue sparing. However, though Flash has been shown very effective in small-animal pilot studies, it has not been implemented to treat human patients because corresponding treatment systems and dosimeters that can accommodate the UHDR are not ready clinically. To address the latter challenge, we will develop OSLD-FLASH, a robust dosimetric system for Flash using sensing approaches for high doses and UHDR - optically stimulated luminescence (OSL) dosimetry, where trapped charges are released by photoionisation to provide ultrafast luminescence in proportion to the exposed dose.

RESEARCH APPROACH: Previously we worked with Defense Research Canada to create an OSL dosimetry system for border security using nanofabricated avalanche photodiodes focusing. In this proposal, we focus on Flash and plan to explore two novel OSL dosimeter (OSLD) approaches to address this need - firstly, novel passive nanostructured materials suitable for fast high dose OSLD, and all electrical nanostructured charge sensors. In the latter case, we will leverage our previous discovery published in Nature Nano for single nanowire (NW) electrometers that can achieve sensitivities better than 10-5 e/?Hz in a 30 nm diameter InAs NWs. The approach takes advantage of weak screening in 1d and such NW systems could be used as sensors for single molecules in gas or aqueous phase. In the former passive materials, we would explore previously developed templated quantum dot systems and for optimised ultrafast radiative recombination on photoionisation. The combination of approaches should identify a novel and most promising route for a novel OSLD for Flash.

NOVELTY AND EXPECTED SIGNIFICANCE: Clinical translation of Flash is extremely risky in patient dosimetry because of UHDR. OSL as a new technology providing dose-rate independence is ideal for Flash. This impactful project providing a dosimetry system for Flash enables clinical translation, offering an innovative RT modularity that can greatly improve the cancer patient survival rate, and potentially increase the RT capacity by ~25-fold.

The compassionate conservation movement, combines conservation biology and animal ethics, prioritizing non-invasive monitoring, particularly for species at risk. The collection of environmental (e)DNA is a rapidly expanding technological innovation for ecological sampling which meets the non-invasive criteria. But are there hidden ethical trade-offs in this approach? Collecting airborne eDNA is an unexplored and high-risk investigative area in emerging eDNA technology and the focus for this novel project in ethical ecology.

This project will attempt to adapt the novel discovery of airborne eDNA to diagnose animal health with the potential for far-reaching impacts. Animal eDNA has only recently been captured in air for the first time making our approach high risk. We will employ untested prototypes for eDNA collection while simultaneously assessing the ethical implications of the technology and potential for mis-use.

Our interdisciplinary approach combines genomics, animal ethics and alternative information dissemination in a novel integration of disciplines. We apply this high-risk approach to a model apiculture system to evaluate our ability to diagnose beehive health. We employ a novel video logs (VLOGs) approach to disseminate the work in real time, framing the research for public dialogue in advance of traditional publication. This approach addresses critical aspects of EDI reducing barriers to access, making field work accessible and embracing multi-modal communication, encouraging discourse from multiple perspectives. Our team includes members who are female, early career, have caring responsibilities, visible minorities and members with known disabilities further broadening our EDI approach to project governance.

We will develop this high-risk technology and challenge the paradigm in compassionate conservation.  Are non-invasive methods ethically superior? This project will evaluate the viability of this new technology and enhance our understanding of critical ethical issues by integrating fields during the earliest stages of development. The technology to gather plant and animal eDNA in air represents an exceptionally high reward opportunity in terrestrial ecology but may have negative ethical outcomes that require new regulatory approaches, spawn new commercial components and modifications to public policy. We will push these boundaries while creating ethical guidance for implementation.

Kidney failure is an increasingly common medical condition for which most patients require life-sustaining hemodialysis treatments 3 times per week for 4 hours. Intradialytic hypotension (IDH), a significant and symptomatic drop in blood pressure (BP), complicates ~20% of treatments. IDH results in recurrent end-organ ischemia, is linked to increased morbidity and mortality, and negatively impacts patients' quality-of-life. Patients undergo repeated brachial BP measurements during hemodialysis to monitor for IDH. Once detected, various treatments are initiated to mitigate IDH and its consequences. Unfortunately, current methods of monitoring always result in a significant delay between the onset of IDH and its treatment. Repeated BP measurements are also intrusive and uncomfortable.

We seek to combine clinical research expertise related to IDH with biomedical engineering expertise related to remote vital signs monitoring and signals processing, to develop novel methods to non-invasively detect IDH reliably and more-rapidly than is currently possible.

Recent research in this area has focused on estimating BP based on time information. For example, proximal and distal sensors are used to monitor the pulse transit or arrival times with BP then estimated using pulse decomposition analysis. The main disadvantage of this approach is that the variables used for mathematical modeling of BP need frequent recalibration to account for changes in vascular tone.

Our proposed research aims to take advantage of recent innovations in the technology and effectiveness of quantum sensors and small wearable non-invasive sensors. Sensors will be deployed to monitor, in real time, variables known to affect BP such as electrocardiogram waves, vital signs, hematocrit and viscosity levels, and other hemodynamic variables in patients undergoing hemodialysis. Neural networking and machine learning techniques will utilize the output of the sensors to construct predictive models for BP variations on a continuous time basis. The main novelty of the proposed research is in associating the predictive BP formulation with the cardiovascular systems of individual patients.

If effective, this novel method for non-invasive continuous cardiovascular monitoring during hemodialysis would allow for nearly-instantaneous detection and treatment of IDH (with important clinical benefits on that basis) while also eliminating the burden of frequent BP cuff measurements during treatments.

Understanding women's reproductive physiology and health requires integration of research ideas and methods from clinical medicine and the study of human adaptation and evolution.  Endometriosis and polycystic ovary syndrome (PCOS) are common disorders that reduce fertility. Polycystic ovary syndrome has recently been shown to be strongly linked with low levels and activity of brown adipose tissue (BAT). Low BAT in PCOS contributes to high body mass index (BMI), insulin resistance, and ovarian dysfunction, as indicated in human studies and by amelioration of PCOS symptoms by BAT transplantation in animals.

A set of recent evolutionary-medical studies has also shown that PCOS and endometriosis represent `diametric disorders', such that they show opposite deviations from controls for a broad suite of traits including BMI, pain sensitivity, follicle recruitment, adult testosterone, and prenatal testosterone as indexed by anogenital distance, AGD.  Prenatal testosterone may thus programme women's developing reproductive systems.

Taken together these findings strongly suggest that there exists a continuous spectrum in reproductive function in women, from PCOS, to healthy normality, and to endometriosis.  And low levels of brown adipose tissue (BAT), programmed by high prenatal testosterone, play a key role in PCOS. The key question then becomes, do healthy women who developed under low prenatal testosterone, and women with endometriosis, exhibit high BAT levels, that contribute to their lower BMI, higher levels of pain, and other endometriosis-related traits?  Levels of BAT have never previously been measured in women with endometriosis, or in relation to prenatal testosterone, pain, or other traits that characterize the spectrum from PCOS, to typical, to endometriosis.  Activity of BAT is also known to vary in relation to ethnicity (the climate to which one's ancestors were adapted), with important implications for evolution and health in this regard.

This project will involve measuring BAT activity (via thermography), and a suite of traits associated with PCOS and endometriosis, in a large set of healthy women (to characterize adaptive human covariation in the traits), and in women with PCOS or endometriosis (to characterize the roles of BAT in these extreme conditions). This proposed work is highly novel and will have profound implications for understanding the reproductive physiology of typical women, and the causes of endometriosis and PCOS.

As the building stock within community housing providers' portfolios age, with many built pre-1980, it is imperative that retrofit solutions are developed that address the specific challenges that they face. These challenges include the limited availability of community housing, the inability to displace residents during retrofits, the vulnerable nature of the tenants, and a current lack of data associated with occupant comfort, health and energy consumption. As such, this research program brings together experts in engineering, architecture, and environmental economics to develop social welfare enhancing solutions for comprehensive, climate resilient, deep energy retrofits that meet the needs of both Canadian community housing providers and tenants.

To achieve this overall goal, three objectives have been developed, each of which will form its own distinct, but interconnected project. These are:

� The design, modelling and experimental evaluation of deep energy retrofit solutions for buildings within Part 9 (low rise residential) and Part 3 (high rise and multi-unit residential buildings) of the building code that can be implemented with minimal displacement and negative impact to the occupants.

� The development of methodologies to increase building resiliency and adaptation against changing climactic conditions, including increasing temperatures and frequency of extreme weather events. Retrofits will be developed to improve building resiliency and designed to help mitigate negative indoor environmental conditions that may result from a loss of services/power after extreme events.

� The development of econometric models to determine the social benefit of the building retrofits on a macro scale (housing provider portfolio) and at the unit and occupant level.

This work is critical to society ensuring the most vulnerable members of the Canadian community have access to safe, resilient, comfortable and energy efficient housing. It is expected that the research program's outcomes will also influence large scale housing portfolios and institutional buildings, including long-term care and retirement homes. This proposed research is timely as there is an urgent need for action in addressing the aging Canadian housing stock with climate change resilient solutions. Without the proper intervention, older Canadian buildings will continue to deteriorate, drastically reducing their livability and affordability, leading to poor housing conditions for occupants.

α-Olefins are fossil-fuel building blocks on which society relies for essential molecules, materials, and pharmaceuticals. As we near the end of the petrochemical era, sustainable access to α-olefins is urgent. Plant lipids have been intensively studied as a renewable source of α-olefins, but the demand for farmland is a severe limitation. Aquatic microalgae, like plants, produce biomass from light and atmospheric CO2, but without need for arable land. The use of microalgae for biofuels production, intensively studied a decade ago, foundered under the enormous freshwater demands for cultivation, and difficulties in achieving sufficiently high biomass. A fundamental problem is balancing biomass vs lipid accumulation: stressful conditions promote lipid production, but limit growth.

Extremophilic algae offer a potent, unexplored alternative. Cold-water extremophiles are unmatched for their ability to thrive in conditions often untenable to conventional algae: poor light, low temperature, and high salinity (which obviates freshwater needs). Lipid biosynthesis is promoted by these conditions as a natural adaptive mechanism. The result is robust growth and high lipid levels, without sacrificing biomass. We will focus on the green alga Chlamydomonas priscuii, one of the best characterized cold-water algal extremophiles, which displays robust growth and high triglyceride levels at high salinity and low light. Our aim is to use these algae as living bioreactors to generate α-olefins via non-natural, bioorthogonal molecular catalysis.

We will employ olefin metathesis (an invaluable, Nobel Prize-winning methodology in organic chemistry) for in-cell remodeling of algal unsaturated fatty acids into α-olefins. Olefin metathesis, despite enormous potential, has been held back by catalyst sensitivity to water, oxygen, and polar nucleophiles. Hence, olefin metathesis in living cells has barely been examined. Chlamydomonas reinhardtii (a model organism and warm-water relative of C. priscuii) is the sole example in microalgae: the catalyst successfully reacted with ethylene in living cells, but not with triglycerides, implying catalysis decomposition. We will use novel catalysts developed to withstand air, water, and nucleophiles, to pursue in-cell metathesis in lipid-rich microalgae. The release of α-olefin products from living cells would open the door to continuous bioprocessing, with transformative implications for the sustainable production of essential chemicals.

Biofilms represent a serious problem in human health, being responsible for antimicrobial resistance, chronic infections, and contamination of medical devices, and are thus a public health priority. Biofilms are a mixture of bacterial, fungal and viral particles, although each microbe are usually studied individually. While it is completely counterintuitive to investigate the interactions between mammalian viruses and bacteria, emerging data suggest that viruses can modulate (promote or disperse) biofilm formation. Bacterial-viral interactions within a biofilm represent an unexplored and challenging area. Our main objective is to understand and control both the microbiological and mechanical properties of polymicrobial biofilms, by combining complementary expertise in microbiology, (bio)material science and engineering. We will: 1) Determine biofilm composition, organization, and the importance of microbial partners; 2) Monitor quantitatively the dynamics of biofilm formation using both bulk and interfacial rheology; 3) Determine the impact of antimicrobial agents on biofilm integrity and microbial survival, infectivity, and stability. Under the guidance of our clinician co-applicant, we will develop model systems relevant to clinical biofilms wherein mammalian viruses will be added at different times of biofilm formation. Fate and persistence of the microbial communities (bacterial and viral) will be determined by biomass, viability, microstructure, and compositional analysis, with and without antimicrobial components, such as our synthesized nanoparticles. Rheometric analyses, that quantify the viscoelastic (i.e. mechanical) properties of materials, will be used to monitor polymicrobial biofilm formation kinetics and disruption in real-time. Clinical biofilms isolated from medical devices will be characterized by the protocols developed with the biofilm models. The multidisciplinary approaches and expertise of the team will provide insight on formation and disruption of biofilms by using antimicrobial agents. This project will lead to new mechanistic understanding of polymicrobial biofilms and strategies toward the urgently needed development of novel antimicrobial approaches. Undertaking this complex task requires an integrated range of multi-disciplinary expertise in clinical microbiology, bacteriology, virology, biomaterial science and engineering.

Insect (arthropod) herbivory has damaging consequences on Canadian agriculture, reducing yield and quality. Pre-harvest losses due to arthropod pests are estimated at 13-16%, rising during outbreaks. Climate change is exacerbating these conditions, expanding the reach and intensity of current and novel pests. The conventional approach uses overwhelming amounts of pesticides as a broad and blunt tool to eliminate threats to agriculture. However, there are compounds known as synergists that are not insecticidal but, rather, inhibit insect detoxification systems, thereby increasing pesticide efficiency. Synergists used in combination with pesticides can reduce the usage of the latter. However, a lack of incentivization has driven chemical companies to focus instead on identifying new pesticides. We propose the identification of synergists that inhibit a key detoxification enzyme of the pest: the glutathione S-transferases (GSTs). Research shows that the delta and epsilon classes of GSTs are unique to arthropod pests and are excellent targets for inhibition by synergists. We hypothesize that plants have evolved chemical defences against herbivory, including inhibitors of insect detoxification enzymes; we can coopt such compounds as biosynergists. Objectives & methods: I) Identify plant-derived biosynergists: genes encoding GSTs from target pests (e.g., Colorado potato beetle) and non-target insects (e.g., bees) will be synthesized. Compounds from a plant natural product library will be screened against the GSTs to determine if they inhibit its detoxification ability. II) Study synergist-GST interactions at a molecular level using high-resolution biomolecular mass spectrometry to characterize inhibitory action. III) Deploy biosynergists along with approved pesticides on (non-)target species: candidate compounds will be encased in silica-based nanomaterials and applied to test species. Novelty/Significance: Research on synergists has not been prioritized by the ag-industry and is limited to decades-old compounds that indiscriminately inhibit pest and non-pest detox enzymes. This project combines complementary expertise in plant metabolism, insect physiology, biophysics and chemical biology, leveraging lessons from human drug development to implement a paradigm shift in mitigating insect herbivory. We are deploying these novel strategies with the help of targeted nanotechnology to enhance pesticide efficacy while ensuring food security. 

In Canada, 84% of post-mastectomy reconstructions are performed using silicone breast implants. Most of these require revision or removal after two decades due to the formation of a thick, painful, fibrous reaction around the implant produced by abnormal activation of the immune system. Reduction of so-called capsular contracture has been achieved with surface texturing, but recent data demonstrates this technique is associated with the development of anaplastic large cell lymphoma in up to 1:1500 patients that has lead to the withdrawal of these implants from clinical use. Thus, there is an urgent need to engineer strategies to combat severe fibrous.

Studies have identified critical roles of macrophages and fibroblasts in fibrosis around implants. These cells can perform diverse functions; the degree of fibroblast activation defines fibrous deposition, and macrophages can both drive and modulate this behaviour. Recent advances show this behaviour is controlled by the way macrophages make their energy. Prior pharmacological approaches to modulate macrophage and fibroblast activities have been limited by targeting individual pro-fibrotic activities with uncontrolled sustained drug delivery that fails to account for the functional changes and communication between these cell types. This project looks to combine novel controlled release approaches with recent mechanistic discoveries to minimize fibrosis.

Our objective is to build polymer materials with small molecules that target macrophage energy systems to control inflammation, paired with anti-fibrotic pharmaceuticals which target fibroblasts via linkages that respond to target undesired cell function. This approach is a departure from previous coating strategies; the coating material building blocks are the therapy, and will apply knowledge in precision polymer development to achieve controlled and tunable release.

To achieve this, a plastic surgeon and polymer engineer will collaborate to: (1) identify design targets for clinical application; (2) build core polymer materials that can instruct macrophage inflammation when they breakdown; (3) conjugate fibroblast-targeting drugs using a macrophage behaviour sensitive linkage to provide on-demand release; and (4) assess the impact of novel coatings on periprosthetic fibrosis in a pre-clinical model of capsular contracture. This technology has the potential to augment gold standard implant technologies to improve clinical outcomes in cancer care. 

The ability to quickly adapt cerebral blood flow to local demand has been key to the development of functional brain imaging techniques. However, the mechanisms by which vascular dynamics is regulated during behavior across spatial and temporal scales remains poorly understood. Noradrenaline (NA), produced by locus coeruleus (LC) neurons of the brain stem, promotes several cognitive functions such as learning and behavioral response. LC-NA has been proposed as a vasoconstrictor in the brain, yet, given that LC-NA innervations are present in most brain regions, it remains unclear as to how noradrenergic activity can locally regulate neurovascular activity. Here we propose to verify that spatially heterogeneous release of LC-NA in the cortex contributes to vascular regulation across the cortex of behaving mice. To simultaneously study cortex-wide NA and neurovascular activity, we propose an interdisciplinary approach merging physics, optics and neuroscience for the development of new imaging and analytical tools. Specifically, we will:

Aim 1: Evaluate the synergy between NA and neurovascular activity during behavior. We will design an imaging method to record neurovascular activity and a fluorescent sensor for NA dynamics across the cortex of head-fixed mice performing a learned behavior.

Aim 2: Determine the role of NA in modulating cortex-wide neurovascular activity. We will engineer our system to include optogenetic stimulation/inhibition of LC noradrenergic neurons.

Neurovascular coupling and neuromodulator dynamics are typically studied at the cellular or micro-network levels, in separate experiments, and in anesthetized animals preventing us from detecting their interaction at the brain-wide level during complex behaviors. Moreover, blood flow and oxygenation level along with NA dynamics are technically challenging to measure with current imaging methods and generate large datasets of complex spatio-temporal dynamics. Thus, measuring noradrenergic modulation of neurovascular activity requires a combination of advanced optical techniques, mathematical methods, and behavioral neuroscience. By bringing together a physics and a neuroscience research team, we will unveil critical neuromodulatory mechanisms responsible for controlling neurovascular coupling. These results could influence the development of new experimental procedures for functional imaging and inform the study of neurodegenerative diseases implicating neuro-vascular interactions. 

The act of scrolling through social media posts is now a near-ubiquitous practice among Canadian youth and young adults. Corroborated by the 2018 Canadian Internet Use Survey, social media is regularly accessed by about 9 in 10 Canadians aged 15 to 34. As one of the singular means to maintain contact with extended friends and family, to obtain fast, easy, and convenient access to personalized news, entertainment, and shopping, and even to generate additional income, the benefits of social media for young Canadians are evident.

The most influential social media platforms are corporate entities that seek to profit through the massive collection of marketing data and algorithmic strategies to influence consumer behaviour. While these algorithms are efficient business tools, they can have harmful societal impacts. Recent studies have explored the link between social media with psychological stressors such as anxiety, depression, eating disorders, suicide, self-image distortion and self-harm. Some have compared social media use to drug and tobacco addiction, and parallels might be drawn between the days of warning-less cigarette packaging and no government-mandated legal drinking age, and the current lack of robust government regulation of social media platforms.

The goal of this study is to ignite the debate on social media platform governance from a new perspective: digital citizenship. Young Canadians learn digital citizenship skills at home and school. However, digital platforms play an important role, since the way students learn is also shaped by their experiences online. We identify the conditions under which social media is a positive or negative force for youth to practice how to: interact ethically, and safely online; cultivate and manage their digital identity; exercise rights and obligations; and maintain digital privacy and security. This study brings together disciplines in ways that have not been previously explored and from perspectives of psychology, social cybersecurity, network science, and artificial intelligence. Our methodology investigates the impact of social media algorithmic strategies on citizenship behaviour and weds aspects of behavioural studies, social influence theory, algorithmic bias, privacy-preserving methods, and deep learning models. This study will challenge the current social media practices and empower young Canadians to develop the knowledge, skills, and competencies to play a full part in all aspects of society.

Planets form in protoplanetary disks surrounding newborn stars. Forming planets in disks can be identified and characterized by comparing observed disk structures with those produced by disk-planet interactions in numerical simulations. Such planets are crucial to deciphering how planets form. A key bottleneck is that a huge amount of computational and human resources is needed to evolve a disk-planet system in simulations. These simulations consume a huge amount of electricity on supercomputers, are expensive, and generate a lot of CO2 emissions.

Our objective is to develop a deep learning method that can efficiently solve the hydrodynamics in disks, and enable disk-planet interaction predictions in a matter of seconds instead of hours or days as we do currently. We will base our method on two new techniques: Physics Informed Neural Network (PINN), which trains neural networks using physical laws, and Deep Operator Network (DeepONet), a new deep network design specializing in learning nonlinear operators in dynamic systems. We will further combine our deep learning method with a Markov Chain Monte Carlo sampler to develop a fully automated package to identify and characterize planets.

The project will be of high reward. Our PINN-based tools will enable fast process of disk observations to identify and characterize planets, thanks to the orders of magnitude improvement in speed and their automated nature that relieves the need for intensive human interventions. This will enable a large-scale statistical analysis of planet formation in disks for the first time, and provide the most direct and quantitative tests to planet formation theories to date. Furthermore, our work will have massive ethical impacts related to the climate crisis. 200 peer-reviewed papers performing hydrodynamic simulations of protoplanetary disks were published in 2021 alone. These works consumed 3,000,000 kwh of electricity, and resulted in 1,200 tons of CO2 emission. Our tools will reduce those by orders of magnitude.

Meanwhile, our program will be of high risk. Our approach of applying deep learning to replace simulations in protoplanetary disks is completely novel. As new techniques popularized in just the past few years, PINN and DeepONet have witnessed tremendous developments, but not in solving equations as complex as hydrodynamics in disks, which are governed by time-dependent compressible Navier-Stokes equations. While adventurous, we are ready to face challenges.

Human health depends on the proper folding of proteins. For instance, fatal diseases like Alzheimer's, Parkinson's, and prion diseases are due to aggregation of misfolded proteins in the brain and other tissues. Consequently, tremendous effort is being devoted toward predicting the folding and misfolding pathways of proteins to better understand what induces a protein to misfold and how misfolding can be avoided or possibly reversed. The folding pathway of a protein (FPoP) and its folding intermediates is determined most precisely through molecular dynamics simulations (MDS). Unfortunately, current state-of-the-art MDS are unlikely to provide the FP of proteins (FPoPs) much longer than ~100 amino acids (aa's) over more than a few milliseconds. Considering that the median length of proteins is greater than 300 aa's and proteins can fold over tens of minutes, radically new procedures must be introduced to reduce the number crunching required by MDS to predict the FPoPs. One particularly effective mathematical tool used currently to reduce computing time consists of running MDS favoring contacts between aa's that are native contacts in the known protein structure. In contrast, this proposal aims to reduce computing time by using parameters, obtained by fluorescence experiments, that describe the motion of aa's in unstructured long chain polypeptides (uLCPPD) made of up to four different aa's. uLCPPD are minimalist representations of proteins, which are made of at least 20 different aa's. The utter simplicity of these polypeptides, which no biologist would think of studying, given how remote their composition is from that of real proteins, allows polymer chemists (PC) to draw rules based on aa sizes, which control the motion of the aa's of a protein as it folds in solution. These rules can then be applied to predict the loci in the protein sequence, where folding occurs. The compartmentalization of a protein sequence into domains significantly reduces the number of aa's contacts to be considered in the FPoPs.

The proposed approach brings unlikely disciplines together by bridging the simplistic uLCPPD studied by PC with the structured proteins considered by biologists. It is risk-taking, for it uses information obtained from uLCPPD instead of a protein structure to determine the FPoPs. By guiding the early stages of protein folding, it could enable MDS to calculate the entire FP of any protein, an accomplishment, that still remains an elusive goal.

The discovery of inhibitory immune checkpoints (ICs) has enabled the development of breakthrough therapeutics that block these pathways and augment immune activation. ICs are generally considered to be protein:protein interactions between different immune cells, for example PD-1:PD-L1 - relevant in T cell-mediated immunity. In this case, ICs assemble between T cells and antigen presenting cells (APCs) via a cell:cell interface called the immunological synapse (IS), which is central to adaptive immune responses. Contemporary work has shown that proteinaceous ligand:receptor pairs are not the only interactions at an IS that can influence immunological outcomes. We now know that cell-surface glycans can interact with immunomodulatory lectin binding partners and influence immune activation through pathways that are mechanistically distinct from protein:protein-based ICs. Progress in this space has been limited by a lack of robust tools to discover and mechanistically dissect biologically relevant glycan:lectin interactions - the glyco-immunointeractome (GIA). This is due to intrinsic challenges in studying glycans which, unlike nucleic acids and proteins, are not encoded by a genetic template. New technologies to better illuminate the GIA in the context of immune cell physiology will advance progress in this space and reveal the fundamental glycan-mediated processes that influence immunological outcomes.

Here, we will develop an unbiased GIA profiling platform using a recently developed glyco-engineering technology to characterize novel glyco-immune checkpoint molecules at the T cell:APC IS. Primary immune cell surfaces will be glyco-engineered using exo-enzymatic labelling with glycosyltransferases to selectively install synthetic derivatives of sialic acid, a sugar often found at the termini of mammalian glycans. The sialic acid derivatives will contain a small photo-activatable probe, enabling covalent capture of the GIA binding complex. UV-irradiation will induce covalent crosslinking of glycan:lectin binding partners between T cells and APCs and be followed by downstream processing of cells to release these conjugates. We will then perform deep glycoproteomic profiling experiments. Combined, this workflow will inform functional immunoassays and elucidate novel glyco-immune checkpoint targets to better understand the roles of glycan-mediated interactions in cellular physiology.

Osteoarthritis (OA) is a chronic disease of the joints that affects 7% of the world population. There are currently no cures nor approved disease-modifying drug therapies for OA. OA has traditionally been dichotomized into disease, characterized by structural degeneration, versus illness, characterized by symptom presentation. This dichotomy fails to capture a holistic representation of OA in the individual patient and highlights an outstanding need for deep phenotyping of OA.

Capitalizing on the Osteoarthritis Initiative (OAI), a prospective study of knee OA with over 10 years of follow-up data from over 4,000 participants that includes clinicodemographics, imaging, biospecimens, and genetic data, our objectives are to: 1) develop and validate a novel latent quantitative trait (LQT) that combines structural and symptomatic features of OA over time [9 months],  2) perform integrative genetic (genotyping array) and epigenetic (microRNA-sequencing) analyses to characterize the unique epi/genetic features of the LQT [9 months], and 3) identify established OA correlates as potential mediators between the epi/genetic features and OA symptoms [6 months].

Approach: 1) We will develop Bayesian multivariate latent variable models that combine both structural and symptom OA features into a continuous progression phenotype (i.e., LQT), which captures the irreversible nature of OA by following a non-decreasing disease trajectory process. 2) We will perform transcriptome-wide and colocalization analyses that integrate genetic data and microRNA-sequencing to pinpoint variants and microRNAs that may be associated with the LQT. 3) Pinpointed variants and microRNAs will then be analyzed in a causal mediation framework to determine direct and indirect effects mediated through other correlates of OA (e.g., muscle properties).

Novelty: Establishing a validated LQT that integrates changes to OA structure and symptoms over time will define disease progression as a continuum which provides a granular representation of the disease condition and will account for the deteriorating trajectory of OA by assuming the LQT is governed by a monotonic non-decreasing process.

Significance: Identifying genetic risk loci and microRNAs associated with the LQT that either separately or together characterize the OA phenotype will help in uncovering biologically relevant mechanisms that act in either/both structural or/and symptomatic traits and will serve in identifying therapeutic targets.

Background: Organ transplantation is the optimal and lifesaving treatment for patients with end-stage organ disease. Although organ transplantation improves patients' survival, they experience grief related to graft rejection, death of fellow patients and death of the organ donor. Grief is associated with failure and loss of imagined past and imagined future. Transplant patients are encouraged by society to be grateful for the organ donation but there is little attention to the difficult feelings that patients and health professionals could experience. This lack of acknowledgement could lead to "disenfranchised grief". Storytelling and creative co-writing have been suggested to support patients, families and health professional experiencing grief and loss after transplantation. However, there are no report of these interventions having been implemented to support grieving transplant patients, donors, families and healthcare professionals.

Objective: The objective of this study is to better document the kidney transplant patients' experiences of grief and loss and to co-develop, with health professionals and writers, writings related to grief and loss that will be posted on our web platform, so as to foster and further understanding of this specific grief in the healthcare community.

Methods: We will first conduct focus groups and interviews with kidney transplant patients, caregivers and health professionals from two transplant programs (Université de Montréal and McGill University) in order to better understand their grief experiences. Secondly, the community of "grievers" will co-create stories and fictions related to death and the imagined past and future. These productions will be posted on a web platform so as to gain a wider visibility. Finally, with individual interviews, we will formally assess the impact of participating to creative activities for participants and we will also assess how the artistic creations are perceived by other health professionals, patients and transplant providers.

Anticipated outcome and the novelty of our research: This innovative and transdisciplinary research project has the potential to support the community involved in transplant and experiencing grief and loss after transplantation through creative writings and humanities. This project includes researchers from medicine and literature and will be used to raise awareness for future healthcare professionals about the challenges faced by transplant patients.

Microbes are key drivers of biogeochemical cycles, particularly the nitrogen and carbon cycles. Yet, it is assumed that microbes do not interact with chemicals in the atmosphere despite being components of bioaerosols. This knowledge gap limits the development of novel, sustainable biotechnologies to address anthropogenic atmospheric impacts.

Objectives: The goal of this project is to provide novel insights into the role of microbes in atmospheric chemistry and to examine the case for using atmospheric microbes in environmental biotechnology. Specifically, we will address three objectives: 1) Determine the genetic potential for atmospheric chemical transformation in the aeromicrobiome using metagenomics and innovative cultivation strategies, 2) identify and quantify the transformations of atmospheric chemicals catalyzed by airborne microbes under atmospheric conditions, and 3) define the physiology required for microbes to survive and grow in the atmosphere using mathematical modelling.

Approach: Our multidisciplinary team will combine their expertise in microbiology, atmospheric chemistry, `omics and mathematical modelling to solve outstanding questions in the roles of airborne microbes in atmospheric processes. Our approach involves the use of methane oxidizing microbes (MOs) as model organisms. Certain MOs are well-known to indiscriminately oxidize diverse chemicals with the enzyme methane monooxygenase, and as the only biological sink for methane, MOs are ideal model microbes for this work. We will use microimpinger techniques to collect ambient bioaerosols for metagenomic sequencing and cultivation. Transformation products of MOs will be identified and quantified using online time-of-flight mass spectrometry.

Novelty and significance: Atmospheric pollutants including greenhouse gases have negative impacts on human and ecosystem health and contribute to climate change. Novel strategies to mitigate atmospheric pollutants are urgently needed. This project will provide definitive insight into the interactions between microbes and atmospheric pollutants. We will identify the chemical transformations undertaken by atmospheric microbes, defining the putative role of microbes in atmospheric chemistry.  Our work will investigate and define what is required for microbes to grow under harsh atmospheric conditions, which is necessary to enable the widespread use of atmospheric microbes for greenhouse gas mitigation and atmospheric pollution management.

Antimicrobial resistance threatens the effective prevention and treatment of an ever-increasing range of infections, and in the clinic, diagnosing and treating resistant infections is often done by trial-and-error. Presently, the gold-standard diagnostic for microbial identification includes a combination of culturing and MALDI-TOF. This process takes approximately 24-48 h, and during this time, patients often receive broad-spectrum antibiotics, which may not be needed and/or ineffective and, increase the selection of new microbial resistance in the population. Moreover, these current diagnostic methods lack the specificity to differentiate closely-related microbial species and they cannot detect the presence of antimicrobial resistance within a strain. To improve detection, our team has developed a platform using high specificity and high sensitivity liquid chromatography - tandem mass spectrometry in combination with machine learning models to identify microbial species from clinical samples. For example, using machine learning-defined peptide signatures, we provide a proof-of-concept study to detect and identify 15 bacterial species responsible for Urinary Tract Infections (UTIs) in up to 300 clinical samples per day. Although our new approach is fast and reliable, we are limited in the number of species detected and unable to assess the status of resistance.

In this proposal, we leverage our expertise in the interdisciplinary fields of microbiology, mass spectrometry, machine learning, and bioethics to adapt our platform for a high-risk approach to identify all pathogens causing UTIs (i.e., bacteria, fungi), and to detect the presence of antimicrobial resistance. We hypothesize that the identification of resistance-specific signatures combined with pathogen identification will enable rapid and reliable diagnosis of the infection-causing agent and inform the clinician as to the best drug to use. This ground-breaking approach will enable a clinician to diagnose and properly treat a UTI in a couple of hours. Further, we will expand beyond UTIs and clinical samples from urine and use our innovative combination of mass spectrometry and machine learning models to diagnose and determine proper treatment from blood (e.g., sepsis), tears (e.g., keratitis), saliva (e.g., COVID). If successful, our high-reward approach will revolutionize the treatment of resistant infections and make personalized medicine an accessible reality for clinicians and their patients.

Context: Prostate cancer (PCa) affects 1 out of 7 men. Resistance to hormone therapy leads to the most agressive state of PCa. Identifying the underlying mechanisms of resistance may yield entirely novel treatment strategies. We propose to investigate a recently identified metabolic pathway, the `glycerol shunt', involved in glucose, lipid, and energy metabolism and associated with aggressive PCa. Glycerol shunt is orchestrated by glycerol-3-phosphate (Gro3P) phosphatase (G3PP), whose expression is high in aggressive PCa and in the hypoxic tumor core. We hypothesize that a high rate of glycolysis in hypoxic tumor cells produces lactate and also glycerol via G3PP mediated hydrolysis of glycolysis-derived Gro3P; glycerol and lactate diffuse to the cancer cells in the normoxic areas of the tumor to serve as fuel; and this metabolic cooperation, between hypoxic and normoxic PCa cells, contributes to the aggressiveness of PCa. To validate our hypothesis, the cornerstone of our approach is the use of two new highly realistic 3D cancer models: the microdissected tissue (MDT), a miniature ex vivo tumor model using patient biopsies, and large spheroids which encompass both hypoxic and normoxic regions simultaneously. Both models are cultured in bespoke microfluidic devices for up to two weeks with high viability and active metabolism.

Objectives:

1. Apply two hypoxic tumor-on-chip models for the study of the newly discovered G3PP pathway.

2. Measure metabolite distribution using 3D metabolomics with high spatiotemporal resolution and elaborate a mathematical model describing transport of nutrient and metabolism in tumors.

3. Measure the effect of G3PP expression levels and glycerol shunt to modulate the response of aggressive PCa to radiotherapy and chemotherapy on chip.

Impact: Metabolic cooperation has never been directly proven experimentally in any cancer and microfluidic devices will provide a radically novel approach to address this. Our synergistic and interdisciplinary team combines unique expertise on metabolism, prostate cancer, 3D biology and microfluidics. We anticipate that establishing a direct link between G3PP and treatment resistance will improve our understanding of fundamental PCa cellular metabolism and unveil a new druggable pathway for PCa treatment and likely for many other metabolism-driven diseases. As part of this project, we will train scientists and engineers at the interface of biochemistry, microsystems engineering, and oncology.

While often a life-saving therapy, mechanical ventilation may promote lung damage, a phenomenon called ventilator-induced lung injury (VILI). Acute respiratory distress syndrome (ARDS), a severe lung disease worsened by VILI, afflicts 10% of all patients admitted to an intensive care unit. With an unchanged mortality of 30 to 41% despite multiple large RCTs, new treatment approaches in ARDS are needed. Assessing mechanical properties of the lung is a promising avenue to detect deleterious mechanical ventilation. Excessive pulmonary strain is thought to be a critical factor in VILI development. Unfortunately, pulmonary strain is hard to measure and most techniques measure global pulmonary strain. Areas of higher regional pulmonary strain, often not identified by global pulmonary strain measures and standard monitoring, are associated with pulmonary inflammation. Detecting these may help to prevent VILI.

We propose to combine passive elastography and lung ultrasonography to accomplish 3 objectives: (1) adapt and optimise an existing vascular ultrasound (US) elastography platform for pulmonary use; (2) use this platform to characterize lung US-measured regional pulmonary strain in healthy volunteers; and (3) measure regional pulmonary strain and its association with outcomes in ARDS patients.

Methods/approach: We will train and validate a deep neural network to segment and track the pleura. We will test different elastography estimators and analysis parameters to optimise the elastography platform. We will use this updated platform and generalized linear mixed effect models to establish normative values of lung US-measured pulmonary strain in healthy volunteers. Similarly, we will assess the association between regional pulmonary strain and inflammatory cytokines as surrogate markers of outcome for ARDS patients.

High risk: The lung does not easily lend itself to US examination as the air-filled alveoli below the visceral pleura reflects US leading to artefacts. While preliminary work supports its feasibility, this proposal is a first attempt to systematically study passive US elastography to measure regional lung strain.

High reward: We seek to develop an easy-to-use bedside tool to measure regional pulmonary strain to improve the safe delivery of mechanical ventilation using widely available US equipment. While ARDS patients stand to benefit the most from this proposal, these results will likely be applicable to all mechanically ventilated patients.

This project focuses on unmet health needs and negative healthcare experiences of marginalized communities, including women, racialized communities, people with disabilities, people living in poverty, 2SLGBTQI communities, and people without citizenship or permanent resident status.

We propose a novel interdisciplinary method to explore racist, ableist, and gendered medical interactions. Using critical theory and community-based health narratives, we will develop an Artificial Intelligence (AI) algorithm and train it to detect medical bias and unmet healthcare needs in a large set of Electronic Medical Records (EMRs).

This study will provide a more nuanced understanding of bias in medical interactions; and of biochemical and cellular pathways of illnesses commonly associated with adverse social determinants of health. The project will also serve as proof of concept for a novel method to address bias in artificial intelligence.

First, we will explore existing community-generated data for narratives of negative healthcare interactions, and health complaints often attributed to medically unexplained illnesses (MUIs). Health complaints associated with MUIs are over-represented in marginalized social groups, and often dismissed by the medical community.

Second, we will examine emerging research on the cellular and biochemical mechanisms of illness that can help to explain MUIs and other functional and somatic illnesses. We will investigate connections between these illnesses and adverse working, living and environmental conditions.

Third, we will develop a novel approach to artificial intelligence algorithms by training the AI to detect medical bias and unmet health needs in the EMRs. Using community-generated indicators, we will search the medical records and examine resulting patterns. AI algorithms currently used to search EMRs are based on medical concepts and practices and reinforce existing medical biases.

We will use critical disability studies, gender studies, critical race studies, and Indigenous studies to explore community-based health narratives; to review the scientific literature; and to build, refine, and train our AI ontology. These theoretical approaches will allow us to consider the broad range of social relations that impact health, including inter-generational trauma; gender-based violence; microaggressions in healthcare interactions; institutionalization; and working and living conditions.

Chronic lung diseases are now the third leading cause of death world-wide (WHO report, 2020). In Canada, they are especially prevalent among First Nations, Inuit and Metis populations due to elevated exposure to cigarette and wood smoke, limited access to primary health care services, and the long-term effects of childhood respiratory infections in residential schools.  Chronic lung diseases cause permanent airway narrowing and alter the types of cells that line the air passages. Squamous cells and mucus secreting goblet cells are increased while ciliated cells that transport mucus and inhaled substances out of the lungs are reduced. Airway remodeling causes much of the morbidity in chronic lung diseases and is considered irreversible.

This proposal is to develop a radically different interdisciplinary approach for treating chronic lung diseases that exploits recent advances in molecular genetics, cell biology, biophysics and nanomedicine. Single cell RNA sequencing has recently defined the gene expression patterns for the different airway epithelial cell types in health and disease, but this new knowledge has not yet been translated into novel therapeutics. We will use markers on the apical surface of goblet, ciliated, and secretory cells to target nanocarriers that have been preloaded with mRNAs encoding critical transcription factors. We will test if delivering these transcriptional effectors broadly or to specific cell types promotes self-rehabilitation of the epithelium. The goal is to reverse disease-related remodeling by restoring normal secretory and ciliated cell populations and arresting the development of goblet and squamous cells that predominate in chronic lung disease. The challenges include custom designing nanoparticles that can be loaded with appropriate cargoes, validating particle binding to appropriate cell types in human airway epithelium, and demonstrating improved physiological function after cell trans-differentiation. We will use cells from lung donors of both sexes and different ethnicities for initial proof-of-principle experiments in vitro, then extend the studies to animal disease models for in vivo assessment.

To our knowledge this approach of restoring a healthy lung mucosa by manipulating cell development using nanomedicine technologies has not been attempted previously, however if successful it would provide a completely new strategy for treating intractable chronic lung diseases. 

Connected autonomous mobile robots that interact with humans in warehouses, on roads, for material handling in Industry 4.0, and now increasingly in healthcare settings, are becoming ubiquitous in our lives. These applications are core strengths of Canada' Artificial Intelligence and Cognitive Science research communities. However, navigation of a network of these robots with a cooperative human-autonomy control feature in dynamic environments, is extremely challenging due to sensing limitations, unknown human objectives, computational constraints, and growing model uncertainties. These challenges impact the robots' predictive capacity for decision making while interacting with humans.

The study of human interactive cognition in joint actions with limited information exchange, and consequent collaborative cognitive processes can notably benefit cooperative controls in networked robots, which communicate decision making and spatial information. Designed from an interdisciplinary perspective, this project will bridge theories in human interactive cognition and those in networked control systems by developing novel cooperative control concepts that learn from smaller scale and weakly structured training data.

Bringing together computer and human neural/cognitive sciences, along with control-engineering disciplines, the long-term goal of this project is to develop a game-changing control theory for human-networked robot interaction through co-design of machine learning and distributed algorithms. Over the next two years, we will address core challenges in modeling joint-action cognitive processes and their neural correlates, and will develop a unique cognitive-aided cooperative control paradigm through two objectives co-designed with experts on human perception and performance: 1) Explore human interactive cognition for learning-aided controls; and 2) Develop a robust semantic-aware distributed perception and controls system for networked autonomous systems interacting with humans.

The benefits from this interdisciplinary approach will have broad impacts in developing scalable and computationally efficient human-autonomy cooperative controls; and pinpointing the key cognitive and physiological ingredients for successful information exchange between individuals in complex multi-agent settings, thus opening a new area of discovery on human-networked robot interaction. The team will also train the next generation of innovators for Canada's robotic industry.

Long-segment airway malignancy, trauma, and stenosis lead to substantial morbidity and mortality. In severe cases, tracheal transplantation represents a solution but fails due to inadequate revascularization. Attempts have been made to indirectly revascularize the trachea by wrapping them in composite allografts such as the omentum and fascial flaps. These approaches however, are limited by tissue availability, require multiple surgeries and lead to complications. Direct vascularization using a tissue engineering approach shows promise. Herein, we aim to evaluate the application of engineered vascular composite allografts for revascularization of a long-segment trachea.

We have investigated approaches to generate bioengineered tracheal grafts with functional epithelium, including development of biomimetic bioreactor systems enabling air-liquid-interface culture and maintenance of viable long-segment tracheal grafts with functional epithelium.  Recently, created biological scaffolds via detergent-based enzymatic decellularization of soft tissue flaps including 1) forearm skin flap, 2) latissimus muscle flap from the back, 3) temporoparietal fascia flap from the thigh, and 4) omentum from the abdomen. These flaps are anatomically equivalent to human tissues currently used in reconstructive surgery. Our research has shown that these tissues can be successfully depleted of existing cellular material, with preservation of extracellular architecture and blood vessels channels. Using perfusion-based bioreactor culture, these scaffolds have been re-endothelialized, via the same blood vessel network, with human umbilical cord vascular endothelial cells (HUVECs). We propose to combine our tracheal regeneration approaches and bioengineered flaps to vascularize a native long-segment porcine trachea using endothelial cells derived from adipose stem cells. Specifically, we will wrap a 5cm porcine graft with a re-endothelialized flap and evaluate graft vascularization in a biomimetic tracheal bioreactor system. 

We hypothesize that tissue-engineered vascular networks surrounding the trachea will allow the process of inosculation with the pre-existing native vasculature around the trachea and neoangiogenesis towards the submucosa. The ultimate goal is to enhance tracheal regeneration therapy by creating pre-vascularized viable tracheal grafts. This study will have significant clinical implications for sustainable tracheal transplantation.

Each year in Canada, 8,300 adolescents and young adults (AYAs, people aged between 15-39) are diagnosed with cancer. Cancer in AYAs often coincides with major life transitions such as entrance to post-secondary education or employment, independent living, marriage and partnerships, parenthood, or caring for aging relatives. Further, the new onset of serious illness at this time of life presents unique medical and psychosocial challenges for AYAs, yet the unique needs of this group remain largely unmet by cancer care systems. However, research indicates AYAs have a clear understanding of how their care can be improved and want to play a more active role in doing so.

This participatory research project will engage AYAs in British Columbia (BC) as co-researchers to design and share an immersive theatre experience with cancer care allies (healthcare providers, decision-makers, researchers, community organizations, and supporters) and other AYAs. The immersive theatre experience will 1) capture the unique and diverse experiences of AYA cancer through an intersectional lens; 2) identify opportunities to improve cancer care; 3) encourage dialogue amongst AYAs and cancer care allies; and 4) facilitate meaningful changes to cancer care for AYAs in BC that are grounded in the lived experiences of AYAs.

The proposed research project brings together the disciplines of health sciences and the performing arts with a specific focus on AYA cancer. As such, the research team is comprised of individuals with expertise as participatory researchers, cancer care allies, AYAs with lived experience with cancer, and specialists in immersive theatre and applied performance.

The research project draws on the methodological expertise of the research team and will apply a tripartite methodology (PAR3) that weaves together three distinct yet complementary, methodologies: Patient Activated Research, Participatory Action Research, and Performance as Research.

This research project is high risk because it engages AYA patients in the design of an immersive theatre experience and invites cancer care allies to actively participate. It is high reward given its potential to facilitate meaningful dialogues amongst AYAs and cancer care allies and identify tangible actions to transform cancer care for AYAs in BC and beyond.

Collective cell movements underlie the formation of embryonic structures such as organ primordia.  Although some mechanisms that orient cell movements such as chemotaxis and planar cell polarity signalling have been studied extensively, they do not explain long-range cell movements in vivo.  Physical properties of tissues, such as stiffness and electrical fields, represent plausible orientation cues but have not been rigorously tested.

Our team synergises developmental biology and mechanical engineering to examine mechanisms that guide cell movements.  Using noninvasive light sheet elastography and a transgenic voltage sensor, both of which we recently developed and validated, we identified endogenous gradients of stiffness and voltage within murine embryonic tissues.  Our current data demonstrate that the position of mesodermal cells within a stiffness gradient explains whether they remain in place or crawl collectively.  However, the direction of cell migration and the presence of cell intercalations in soft tissue remain unexplained.  We propose that endogenous electrical gradients orient cell crawling and promote cell intercalations where stiffness is low.

To examine the combined effects of stiffness and electrical gradients in tissues such as the limb bud and neural crest, we will simultaneously map both mechanical and electrical properties while tracking cell movements in 3D.  Overlapping gradients will be correlated with modes and directions of cell movements.  To define their relative contributions, each property will be manipulated by genetic gain and loss of function experiments.  Fibronectin, which underlies mesodermal stiffness, will be conditionally deleted or over-expressed using a transgenic strain we generated.  Electrical gradients will be manipulated using engineered ion channels in transgenic embryos by agonist or antagonist small molecules.  Quantitative data will be used to generate a random walk model with terms for stiffness and electrical gradients.  To test and refine the predictive ability of the resulting equation, we will examine other organ primordia such as the mandibular arch.

This study will define elusive cues that guide the long-range movement of cells.  The experiments are possible because of complementary tools that were developed through transdisciplinary engagement.  The results will have far reaching implications for understanding development and anomalies, and for accelerating tissue regeneration.

In 2016 British Columbia declared a public health emergency in response to unprecedented rates of illicit drug overdose. Despite all efforts, overdose rates have since doubled with overdose deaths far exceeding those due to COVID-19. The overdose crisis is linked to the emergence of highly potent synthetic opioids such as fentanyl in the unregulated drug supply and people not knowing the potency or full composition that they may consume. Drug checking has been piloted as an overdose response in Canada, however the scale and reach remains limited. Barriers to realizing meaningful impacts due to technological and cost challenges remain. Over the past five years, our team of chemists, social workers, computer scientists, nurses, and adult educators have been developing new optical technologies to analyze drug mixtures that are integrated with harm reduction messaging and public health reporting. Despite our advances, one of the major barriers is the availability of appropriate, low-cost, easy-to-use technologies that don't require a trained technician for data interpretation. Such advancements are necessary for drug checking to be broadly available within public health responses to overdose, notably in suburban and rural communities throughout Canada. Our project seeks to implement and evaluate our current test-bed advancements in automated technologies for drug checking. Our goal is the real-world implementation of a drug checking platform, eliminating the need for specially trained technicians at every site by engaging artificial intelligence-driven automation. These technological and service delivery innovations are poised to be transformative as we address the known risks of: ensuring the drug checking platform meets current demands for accurate reporting of the full composition of each substance, providing quantification of the active components to inform potency, and interpreting results within the context of overall results in the local area and with relevant harm reduction information and messaging. To respond to these risks we will systematically implement automated analysis and reporting functions with the goal of piloting a fully-automated custom hardware and software drug checking platform within two years. The intervention will be evaluated according to the health quality dimensions of effectiveness, appropriateness, access, equity, efficiency, and safety as we engage the Knowledge-to-Action cycle throughout the implementation stages.

In recent years, there has been a proliferation of theoretical approaches addressing challenges of agricultural and food production for coming decades. Nonetheless, there is a lack of practice-based projects that can operate from and through an interdisciplinary and experimental methodology. This project is situated at intersection between disciplines, resulting in a multi-faceted, expansive and collaborative approach. It examines the feasibility of production of Symbiotic Culture Of Bacterial Yeast (SCOBY), from renewable biomass, to be used as a feed additive for poultry. It coincides with the production of a film that informs and documents processes. Preliminary communication and testing within artistic, scientific and agricultural fields have gained much interest, a response which has inspired the development of the proposal.

The research coincides with numerous anecdotal DIY enthusiasts' promotion of feeding the SCOBY to their chicken for various health benefits. However, large-scale production of SCOBY as a drop-in type of feed additive for poultry has yet been explored, and the potential pre/probiotic benefits of SCOBY for pullets or layers has remained elusive. Therefore, this project will look into variable conditions and methods of growing and processing SCOBY (including utilization of waste biomass), investigating its nutritive value and its ability to be used as a preventative health strategy in laying hen nutrition, and its environmentally friendly production. On an artistic level, the SCOBY visually and practically juxtaposes dry, inert and efficient poultry feed. The camera will follow SCOBY's fleshy, wet body through these institutions, exposing what is often otherwise unseen. In this way, the film reclaims alienated relationships to cyclical processes of material transformation, and expands cultural imagination on how ecological relationships can look.

The impact of these studies, if successful, could 1) disrupt the blackboxing effect  from research and development level throughout agricultural and food production chain, by binding people to processes, both physically and conceptually; 2) facilitate deeper connection and care for waste that translates to sustainable material with low-economic costs, that could benefit overall chicken health and poultry industry products and create progressive jobs within the industry; and 3) cultivates imaginaries for more complex and sustainable relationships to non-human entities and environment.

Although water supply and sanitation are considered as standard public services, the lack of clean drinking water in many First Nations communities remains a problem in Canada. The Government of Canada is committed to addressing drinking water-related health and safety needs. However, the vast territory and disparate locations of communities make it a huge challenge to set up water treatment infrastructure and provide necessary training and maintenance. Point-of-use water treatment technologies offer an attractive alternative. They can be easily transported, installed, and operated, which imply tremendous cost savings. However, the drawbacks of existing point-of-use technologies include frequent filter replacement, needs for power supply, and multiple stages for enhanced treatment. To address these issues, we attempt to develop a self-powered portable device for simultaneous water disinfection and pollutant removal. Hybrid functional materials will be developed and embedded within a portable container which is powered by water flow and sun light.

The simultaneous water disinfection and pollutant removal will be achieved through electroporation and piezophotocatalysis. Nanowires will be used to damage the outer structure of microbes for water disinfection. Piezophotocatalytic materials will be developed, which will utilize mechanical energy to enhance the degradation of various pollutants. These functional materials will be integrated into a portable device. Multiple energy sources such as gravity-driven flow and manual pressing/shaking will be tested. Prototypes will be developed for various applications, such as personal water bottles, high-capacity tanks for family/community, and modular cartridges.

The developed devices will be applied to First Nations communities which suffered from water quality issues. There will be no need for power supply, filter replacement/disposal, or on-site professional service. It will bring about tremendous benefits to First Nations communities in terms of enhanced drinking water quality, reduced public health issues, decreased cost for water supply, and minimized environmental footprint. The developed devices can intake water from various sources, and can be used for ensuring drinking water safety in First Nations communities, providing clean water for remote trips/sites, supporting industrial/military deployments, and supplying drinking water under emergencies (power outages, contamination events, wildfires, etc.)

HIGH RISK: A million Canadians are expected to be living with dementia by 2030, with a 51% increase in the number of new cases per year. The hospital environment can exacerbate symptoms of anxiety, depression, and emotional distress, which often lead to risky physical and verbal behaviors, resulting in costly and unnecessary injuries for staff and patients. Recent evidence shows technology such as virtual reality (VR), e.g., virtual experience with pet visits, shows promising benefits to improving mood and quality of life for people living with dementia in the community.  However, it is unknown how VR interventions can be best adopted and adapted in hospitals for patients with dementia care, especially for those with COVID-19 visit restrictions and families living far away. Hospitalized patients experience digital inequity; they lack access to technology for non-pharmacological psychosocial interventions. Current VR programs are based on easy-to-reach groups living in the community, they do not match the needs and preferences of patients with dementia in hospitals.

OBJECTIVES:

1)   Establish a novel model in a hospital setting where a VR intervention will be co-developed and tested with patients, families, industry, staff, and decision-makers in the field in real-time to accelerate adoption

2)   Identify barriers and enablers to collaborative development

3)   Evaluate stakeholders' experiences and impact

INTERDISCIPLINARY: The study brings researchers in Nursing, Computer Science, Occupational Therapy, and Engineering to work with patients, families, staff (e.g., nurses, care aides), decision-makers, and industry, to co-research in practice. VR can offer high rewards (e.g., quality of care experience) for a largely invisible, vulnerable, growing population of older people in hospitals. It is impossible to achieve inclusivity and sustainability without the involvement of relevant stakeholders.

HIGH REWARD: Dementia care is a longstanding issue underrecognized in Canada. The principle of "nothing about me without me" will guide this research to include the voices of people with dementia and their caregivers in developing innovations for their own care.  The study leverages digital technology to transform future hospital dementia care that requires collaborative capacity building for meaningful impact. The new embedded research model will enable the co-creation of useful technology tools for good quality dementia care.

Specialized metabolism in plants and microbes is a process of remarkable genetic and biochemical plasticity leading to astonishing diversity of chemical structures that have found broad applications. Yet, accessing new compounds with potent activities for drug discovery is extraordinarily difficult. Introduction of genomics and transcriptomics studies have led to discovery of several biosynthetic pathways of important natural products. Yet, this process still stands on mechanistic knowledge and very low throughput approaches. In an attempt to overcome the current limitations in pathway discovery, we propose ACE2D, a disruptive Synthetic Biology approach that harnesses the power of evolution for self-assembling of biosynthetic pathways across kingdoms by mimicking the natural ecosystem interactions. We will engineer a modular approach in yeast that includes a signal pressure component (M1), a randomized DNA library containing genes related to specialized metabolism under different selection markers (M2), and a bioactivity response component (M3) against the induced pressure. Through iterative rounds of adaptation, only cells acquiring the correct combination of genes in M2 for biosynthesis of compounds (of interest) with M3 activity will be able to survive at increasing pressure in M1, thus enabling pathway self discovery, reconstruction and optimization. Increasing complexity of M2 to whole (meta)transcriptome level or combinatorial transcriptomics will enable assembly of hidden or artificial pathways and hence, production of a myriad of novel bioactive compounds. We will perform computational modeling and data science to predict evolutionary projections of natural or artificial pathways, foresee recombineering rate and identify genetic targets to fine-tune system performance. M3 products will be subjected to high throughput screening to discover activities beyond M1 pressure, such as antioxidant, antibacterial, anti-inflammatory, cytotoxic or anti-Alzheimer's disease (AD). We will apply ACE2D to complementation studies of AD-related molecular markers in yeast, under AD-triggering stress (M1). Repurposing M2 to human genes will provide insights of the mechanism of action of bioactive compounds (M3) that reinstate physiological parameters in AD prevalence conditions. By exploring the dark holes in nature's chemical space, ACE2D will become a game changer in natural products research providing rapid access to a rich source of next-generation pharmaceuticals.

There are two possible responses to the global ecological collapse: 1) prevent, reduce, or reverse the effects of human activity on the climate to allow the planet to recover, or 2) adapt by developing technological systems to replace the ecological services on which we rely. For ex., while much has been done to reverse the loss of pollinator species through conservation strategies, researchers are also studying how to replace biological pollinators with technological ones.

Our work builds upon the premise that art exhibitions drive research by providing environments for critical reflection and meaningful dialogue among objects and their audiences. We will change the narrative of the climate crisis by welcoming people to experience our shared future in a Manufactured Ecosystem. This Manufactured Ecosystem will demonstrate a complete adaptation strategy - in which all ecosystem services are replaced by biomimetic technology. The exhibition will feature all of the existing technologies that can replace ecosystem services in combination with an imagining (from science fiction) of those that remain to be innovated (e.g. the photosynthesizer). Visitors will be able to visit in both physical/virtual spaces, the exhibition will be accessible visually and through auditory recordings, and we will develop a curriculum for the exhibits for educational use in elementary to higher education. The experience will be jarring and act as a stimulus to change the narrative about our climate crisis.

Objective 1: Curate an open access database of existing technological proxies for ecosystem services and identify which technologies don't currently exist

Objective 2: Use science fiction literature and art to imagine the technology to replace the ecosystem services that we have not yet developed

Objective 3: Build a physical and virtual exhibition for people to experience and learn from the Manufactured Ecosystem

Objective 4: Collect the reflections of visitors to these spaces

Earth is reaching its planetary boundaries and the UN has given us 12 years to act. In Canada, we are experiencing the effects of crossing boundaries with respect to climate change, land-systems change, and biosphere integrity, all of which strains the resilience of our systems. If we continue on our current course, humans and the biosphere will change permanently and, critically, at a rate that is unlikely to support healthy adaptation.

Quantum computing has shown promise in providing significant computational speedups compared to classical computing. Promising applications lie in quantum chemistry. Motivated by this, ongoing developments have opened up the possibility of running quantum chemical calculations using near-term quantum computing hardware. Such developments are a first step towards building future general-purpose quantum computing platforms, providing the opportunity to run highly accurate and expensive quantum chemical calculations at a reduced computational cost.

Our research leverages this potential to revolutionize the field of quantum chemistry. It will tackle an unexplored area of research that involves interdisciplinary efforts to hybridize fields like machine learning, quantum chemistry, and quantum computation in developing new efficient computational procedures for calculating various molecular properties. Our work focuses on building general-purpose machine learning models for molecular property predictions currently absent from the literature, exploring the feasibility of quantum enhancement in generating such models, and building analogous quantum machine learning models. Some important research questions are: Can the performance of fast quantum mechanical methods be improved by (quantum) machine learning-based corrections for generating a large volume of accurate quantum training data for deep learning models? How can quantum computing provide computational advantages in data processing and model pre-training stages? Can we observe quantum enhancement via developing quantum analogs of supervised learning and deep learning models?

Answers to these research questions from our work will advance data-driven artificial intelligence toward an unexplored area of chemistry application by using quantum computing. Computational methodologies developed as part of our research will demonstrate the future of quantum computing and machine learning within chemistry. Our work will impact catalysis and materials discovery, and the innovations that emerge from our research will bring valuable new computational tools to a broad community of users in chemistry, physics, and other disciplines.

Wetlands are often forgotten features on the landscape that are drained, paved over, contaminated with pesticides and other chemicals, and otherwise seen as impediments to human progress (agriculture, urban and industrial development). But recent research shows that these waterbodies are biotic engines that enhance biodiversity and productivity and export biomass and nutrients in the bodies of adult aquatic insects, subsidizing the diets of wildlife. How this subsidy indirectly translates to human well-being and economic benefits has never been assessed. We will develop a valuation framework for the indirect effects of aquatic insect export from Prairie Pothole Region wetlands on birds, fish, and other consumers and quantify both beneficial and harmful compounds accumulated in insect tissues, namely concentrations of omega-3 fatty acids and current-use pesticides. To do so, we will use an array of experimental wetland enclosures (ie. limnocorrals) and tracer experiments, to 1) measure the quantity and quality of emerging and migrating insects in intact wetlands and in wetlands exposed to pesticides and 2) test the use of 15N and 13C enrichment for tracing insect-based energy and nutrient flux into surrounding food webs. These measurements will provide the data needed to quantify the ecosystem services (provision of biomass, omega-3 fatty acids) and disservices (concentrations of pesticides and reduction in biomass in response to exposure) that result from insect emergence and migration associated with wetlands that differ in pesticide exposure. In concert with these biological and chemical measurements, the valuation framework will link changes in wetland insects with associated ecosystem services endpoints that affect human well-being such as fishing and biodiversity. We will apply economic valuation methods including recreation demand models and stated preference methods to uncover the non-market value of these ecosystem services and gain a more complete understanding of the value of the 6 million wetlands in the Prairie region. Effects of agriculture and other human activities on wetlands are notoriously hard to measure and assumed negligible, creating a high-risk project. Yet we hypothesize that net environmental and economic benefits will occur when wetlands are left protected from pesticide inputs, allowing the export of a nutritious food source that sustains wildlife populations and makes a novel case for wetland and biodiversity conservation.

Annually, more than one million people around the world experience the loss of a limb, which significantly impacts their daily life. Prosthetic arms have great potential to improve the lives of amputees. However, available prosthetic hands have limited functions and the interface lacks intuitive control. Current state-of-the-art non-invasive prosthesis control systems use pattern recognition techniques driven by surface muscle signals. This requires the user to carefully exert distinct muscle signal patterns to perform different gestures. However, in real-life situations, intact people rarely have to think about their hand gestures when grabbing an object, instead, the fingers and the hand are naturally configured to the proper posture when the hand reaches and touches a target object. Surgery to connect sensors of the artificial hand to the nerve in the amputee's residual arm, e.g., Targeted Muscle Reinnervation, is highly costly in terms of both extra surgical cost and potential infections. Inspired by this fact, we propose a novel bio-inspired natural prosthetic hand control interface via the addition of miniature cameras and tactile sensors to prosthetic hands. These additions will enable the robotic hand to "see" the target and "feel" the environment and the object during reach-and-grasp process, and automatically drive the hand towards grasping with little control effort from the user, i.e., the amputees only need to decide whether to proceed or retrieve the robotic hand.

This project brings together the disciplines of Computer Science, Engineering, Rehabilitation, Kinesiology, and Psychology to achieve the following objectives. Objective 1: enable prosthetic hands with vision and haptic functions utilizing computer vision and tactile sensing techniques. Objective 2: explore best prosthetic hand control strategies to achieve high accuracy movement with minimum control effort from the user.  Objective 3: develop an easy and natural prosthetic hand control interface by fusing multiple inputs from computer vision, touch sensing, and muscle signals. We will attach the developed interface to both commercially available and customized prosthetic hands and test with amputees in collaboration with our local rehabilitation center.

We expect this project to deliver an affordable and easy-to-use natural control interface to decrease the rejection rate of prosthetic hand use in real life, and thus benefit amputees throughout Canada and the world.

Emotion dysregulation is a barrier to effective parenting, and is more likely to occur among families of children with Attention-Deficit/Hyperactivity Disorder (ADHD). ADHD is one of the most common neurodevelopmental disorders, and parents of children with ADHD are more likely to experience ADHD and other psychological disorders, household chaos and conflict, parenting stress, negative parent cognitions, and negative parent behaviours, all of which can detrimentally impact parental emotion regulation. Psychological treatments focus on building parental self-awareness of signs of emotional distress to provide opportunities for regulation and more effective parenting. However, this awareness can be challenging as emotional distress impairs self-awareness, thus leading to parenting when in a state of negative arousal. The objective of this study is to leverage multimodal child and parent technology to facilitate development of real-world-based in-the-moment awareness of emotional arousal in daily lives, to reduce the negative consequences of emotional dysregulation. First, we will develop a smartwatch algorithm for detecting emotion activation among parents of children with ADHD by collecting ecologically-valid in-home longitudinal measures of stress. Parents will complete various questionnaires that examine demographic and contextual variables associated with biosensor data. Parents will then use wearable technology with biosensors and complete daily and repeated measures using scheduled and event-triggered sampling. State-of-the-art data analyses will be employed to examine concurrent and temporal relations between measures of emotion and biosensor data to predict emotion activation thresholds. Secondly, we will develop an application integrating this algorithm into a smartwatch, such that the device can provide a vibration to prompt the wearer to engage in emotion self-regulation strategies when emotion activation is detected. No research has examined wearable technology for emotion regulation among families of children with ADHD. A smartwatch vibration is an immediate, less intrusive, and less stigmatizing way of delivering feedback to a parent who may have difficulty coping with emotional distress. The resulting application would be a novel adjunct intervention to a broader population of those who experience challenges coping with stress. Given the increasing use of wearable technology, such a function may have widespread appeal, uptake, and impact. 

Remotely supervised virtual point-of-care ultrasound (Tele-POCUS) can revolutionize clinical examinations by providing the ability to assess the heart, lungs, and other organs immediately at the bedside. This technologic breakthrough allows for live-streaming of POCUS images from remote regions directly to experts thousands of kilometres away. However, Tele-POCUS, just like other virtual care modalities transforming care in the North, poses new questions and implications beyond its inherent medical advantages. How do we ensure advancements in virtual care won't exacerbate digital divides contributing to existing health inequities? How do we quantify the impact of these technologies on Northern and geographically remote communities? Do virtual modalities such as Tele-POCUS result in more human-to-system disconnection, distrust, and disengagement, or are these limitations outweighed by benefits such as obviating the need to extract patients from their communities for diagnostics and care?

We are a team of psychosocial, medical, and digital experts addressing the implementation, sustainability, and social impact of Tele-POCUS upon geographically remote communities. We aim to train, implement, and evaluate the uptake of POCUS by remote care providers. Our goal is to establish a national Tele-POCUS implementation program couched in terms of 1) evaluating the psychosocial impact of Tele-POCUS on patients and providers with relevance to the community makeup and 2) identifying barriers and facilitators to the uptake of POCUS within a broader context of virtual services. We recognize that populations in these regions have a high proportion of Indigenous peoples and will partake in deliberate community engagement with Indigenous communities to ensure their inclusion in impact assessment. The integration of virtual care modalities such as Tele-POCUS will profoundly disrupt care in remote regions, radically altering current diagnostic practices. Care providers may be hesitant to accept this change and patients may not fully support uptake of further virtual services. However, we are committed to partnering with stakeholders from these regions, including them at the forefront of this collaboration to ensure we are best serving their needs. Lessons learned will have far-reaching value to the broader implication of virtual care and provide a profound and immediate benefit towards developing higher quality of care in remote and low-resource communities.

3D-printed breath sampler for capturing and analyzing exhaled microdroplets

SARS-CoV-2 is transmitted by respiratory shedding via exhaled air, including from asymptomatic patients, fueling the uncontrollable propagation. Current infectious disease diagnosis proceeds via samples obtained from nasal swabbing or blood, which are invasive, and sometimes from saliva. None of the current methods provide a measure of viral shedding, which could help assess risk of transmission. For example, for COVID19 there is significant debate about the thresholds for contagiousness, whether positive PCR test or rapid tests are a good measure of contagion, and when it is safe for people who were positive to return to society. However, there is little data to determine at a collective and at the individual level whether someone might be shedding virus outside of the acute viral phase. Contagiousness for airborne and droplet-transmitted diseases could be effectively measured by sampling the exhaled air and quantifying the infectious load in the breath.

Objective

Our goal is to develop a compact, breath sampler made by 3D printing that can capture and accumulate with high yield droplets and microdroplets in exhaled air.

Research

Aim 1: Design, 3D printing and optimization of a compact microdroplet breath sampler

Aim 2: Microfluidic assay of sampled breath microdroplets

Aim 3: In-line breath sampling and analysis of hospitalized, ventilated patients

Significance

If successful, the breath samples could replace nasal swabbing, blood drawing and saliva sampling for many health and disease conditions, thus simplifying sample collection beyond the one afforded by liquid biopsy. Breath samples could be used remotely for rapid testing and point-of-care diagnosis, as well as for mail-in to central laboratories. An in-line breath sampler in ventilated patients could help collect and analyze pathogen and host indicators of disease progression. Finally, a breath sampler combined with a rapid test could serve as the foundation for measuring contagiousness for diseases propagated by respiratory shedding in a portable format, and help study and mitigate the transmission of infectious diseases such as COVID-19. 

Objectives: The ways in which cortical networks reorganize during early development is almost completely unknown, yet critical to understand how novel behaviours normally arise and how brain function is altered in developmental disorders. Traditional approaches require infants to be still, sleeping, or repeating a specific task in order to monitor brain activity, but such states do not reflect normal behaviour and alter functional brain connectivity. This research will use a novel "naturalistic neuroscience" approach to achieve the following: 1) Assess spontaneous changes in functional connectivity in the cortical networks of freely behaving human infants, 2) Use machine learning to identify functional connectivity patterns associated with typical versus atypical development, and 3) Assess the impact of early intervention on functional brain connectivity in atypically developing infants.

Approach: From 4-18 months, full and preterm infants will be assessed using the Alberta Infant Motor Scale (AIMS) to determine whether infants are on a typical or atypical developmental trajectory. Over the same period, resting state functional near infrared spectroscopy (rsfNIRS) will be used to map spontaneous fluctuations in cortical connectivity in awake infants. Both unsupervised and supervised machine learning algorithms will mine the neural data to extract functional connectivity patterns associated with a typical versus atypical developmental trajectory. Infants following an atypical trajectory will receive intervening treatment from an occupational therapist and the impact on cortical functional connectivity will be compared in infants that do and do not respond to treatment.

Novelty & Impact: Applying rsFNIRS in freely behaving infants is highly risky as it can be difficult to distinguish meaningful neural signals from neural noise. Yet, the creation of machine learning algorithms that can distinguish typical from atypical development based on spontaneous neural activity could lead to clinical applications with earlier diagnoses, individualized therapeutic programs with increased effectiveness, and even non-invasive brain computer interfaces capable of restoring mobility in otherwise untreatable cases. Ultimately, this research could help to reveal how the wiring of the human connectome transforms across multiple stages of human development to enable the emergence of meaningful goal-directed behaviour in both neurotypical and atypical individuals

The noble goal of medicine to deliver healthcare to all persons regardless of ethnicity, race, gender, age, or socioeconomic status, requires a more explicit commitment to the ethical principle of equity, in light of notable discrepancies in care. In the Department of Diagnostic Imaging (DI) at our pediatric hospital, approximately 3000 appointments are scheduled monthly. DI is an important step in disease diagnosis and hence, timely access to its appointments is crucial for optimal care delivery to patients.

Although attempts are made to accommodate patients' availability, clinical urgency and demand often result in patients being scheduled for the first available timeslot. To investigate the existence of inequities in access to DI appointments, we have retrospectively collected and analyzed data from 74000 DI appointments during 2018-2021. Our preliminary results show that there is a relationship between negative appointment experiences (NAEs; i.e., no-shows or long waiting room times) and some of patients' demographics. For example, families are more likely to have a NAE if the patient is older or they are likely from a household with a lower income, a single caregiver, or non-English speakers.

NAEs are not only stressful for families but also have a negative impact on the hospital. When appointments are missed, healthcare resources are lost and an appointment that could be helping a patient goes empty. They can also negatively impact how staff interact with families, further compounding inequities. Thus, there is clearly an opportunity to augment equitable access to DI.

In this research, we will develop and prospectively validate a novel Machine Learning-based predictive tool to identify families at risk of NAE to offer a personalized intervention to increase their ability to access DI, minimizing the likelihood of an actual NAE. We will demonstrate that equity-promoting interventions can enhance system efficiency while improving access to care. We will also evaluate equity metrics including parental satisfaction and utilization of adjunctive supports (e.g., preferred scheduling time).

Our project is high risk because there may be confounding factors to the NAEs which may not be reflected in the available patients' demographics leading to compounding disadvantage. The research is high reward because it has the potential to ensure that disadvantaged children receive healthcare that many currently do not receive due to barriers to accessing care.

Canada's Standing Committee on Public Safety and National Security estimates that 70-80% of police service calls involve a person experiencing a mental health crisis, making the police "de facto front-line mental health workers" and "informal responders of our mental health system". Reports of disproportionate uses of force by police have raised questions as to whether police training has advanced in ways that sufficiently equip officers with the skills to meet these rising demands. The consensus among police trainers across the province is that "reality-based scenario training is the most effective way to train for real-life incidents". Beyond the police service, community intervention teams (e.g. nurses, clinicians, civilian volunteers) can also benefit from such training.

However, such scenario-based training currently requires involvement of live actors, and hence, is not scalable to the extent that is required for our front-line first responders. In this project, the research team will develop a Virtual Reality (VR)-based Training System that can 1. Intelligently assess the behavioral response of the first responder undergoing the training in real-time 2. Adapt the VR scenarios based on the trainee's response by modifying the behavior of the VR characters 3. Provide automated evaluation of training performance during and after the training session. The research team will work towards achieving these 3 research objectives through an interdisciplinary approach. For behavioural response assessment of the trainee, the team will employ machine learning algorithms on data collected during training (e.g. physiological sensor data such as ECG, movement data, eye gaze) to assess behavioural indicators that can be mapped to specific training objectives. These indicators will be utilized to intelligently adapt the behavior of the VR characters, so that each training session can be fine-tuned to target specific training objectives for a particular trainee. Finally, the training performance will be evaluated by adapting some recent systemic methods for de-escalation training performance evaluation developed by the research team.

Such an automated tool for de-escalation training is highly scalable. By integrating the tool in the training curriculum of different agencies, the research is expected to result in increased de-escalation training, and, hence, proportionate response by our front-line responders when responding to mental- health crisis calls.

Monocytes are immune cells that play critical defensive roles. However, their overabundance, aberrant mobilization from the bone marrow and recruitment to target organs lead to chronic inflammation that significantly damages the musculoskeletal, cardiovascular and nervous systems, and cardiac muscle after heart attacks.

New paradigm: combine the disparate fields of organic chemistry, vascular disease, nanotechnology and leukocyte biology to radically enhance our understanding of chronic inflammation and challenge the current paradigm of treating the issue with steroids, antibodies and small molecule drugs. We'll develop novel RNA nanotechnology to address a completely new therapeutic target: the bone marrow.

New theory: on-demand silencing of CSF-1 and MCP-1 genes in the bone marrow will slow monocyte generation and their entry into blood circulation, respectively. Permanent silencing would be detrimental to the body's long-term injury and disease response, so the required technology must incorporate on-demand functionality. As preliminary work, we created a synthetic RNA nanoparticle that delivers 2.5x more RNA per nanoparticle, as compared to best-in-class current clinical products used to treat liver disease. Critically, our system also delivers to the bone.

Aims:

1) Formulate our nanoparticles with small interfering RNA (siRNA) that silence the CSF-1 and MCP-1 genes, and show the ability to alter monocyte numbers and populations in the blood of healthy mice. Nanoparticles are administered intravenously and blood analyzed by flow cytometry. The ratio of the two siRNAs will be optimized.

2) Determine optimal nanoparticle dosing regimen. Gene silencing by siRNA is temporary. Via flow cytometry, we determine the time it takes monocyte levels to return to normal after a single treatment. This establishes the duration of the effect and informs the time interval between therapeutic doses.

3) For therapeutic proof-of-concept, an existing well-characterized mouse model of elevated monocytes is used. Using our initial dose and interval, we optimize nanoparticle performance in diseased mice, as inflammation can impact nanoparticle uptake and RNA utilization.

Instead of incrementally improving the current standards of care, we leverage multiple highly unconventional therapeutic targets and a novel, highly efficient RNA nanoparticle system to invent an on-demand treatment modality that addresses the complex challenge of chronic inflammation.

Food security is an emergent global issue given the continual increase in human population (today at 8 billion and trending to 10 billion by 2050), the capricious weather system and exacerbating geopolitical tensions. Hydroponic greenhouses have steadily evolved into a viable means of farming with water utilization efficiency of greater than 90%, crop productivity markedly increased and autonomous local farming becoming feasible.

However, loss in global crop productivity of 20 to 40% annually due to pathogens, pests and weeds is jeopardizing the success of indoor farming. For example, tomato brown rugose fruit virus, detected in Jordan Valley and Israel in 2014 and which has since spread globally, has caused crop loss of 30% to 70% annually. Another example is the bacterial pathogen pseudomonas syringe pv. actinidiae responsible for the severe occurrence of bacterial canker in kiwifruit vines. Consequently, early-stage detection of pathogens in plants is critical to prevent their spread across greenhouses.

In hydroponic greenhouses pathogens transmit rapidly and can be present in hydroponic cultures for extended periods of time (up to a year) prior to onset of disease expression. Hence preventing spread of diseases requires the development of ultra-sensitive and rapid pathogen detection at the point-of-plant-care (PoPC). Current pathogen monitoring practices lack the high-level of sensitivity and reliability - key factors for screening vigilance and efficiency - and are labour intensive and costly. Accordingly, there is a need for an economic highly-sensitive early-stage system for incipient pathogens in hydroponic greenhouses.

We propose the development of a real-time PoPC pathogen contamination sensing system based on innovative highly-reproducible nanogratings which use near-field surface enhanced Raman spectroscopy (SERS) - a highly sensitive molecular finger-print technique that can detect and differentiate biochemical species. The proposed research is a unique confluence of three interdisciplinary advances: chemical synthesis for accurate sample preparation and delivery, multi-wavelength SERS capable of attomolar sensitivity, and artificial intelligence algorithmic spectral analysis techniques. Successful implementation of the envisaged plant pathogen detection system will permit unprecedented detection of incipient pathogenic species, making it possible to mitigate growth and proliferation of diseases in plants.

Greater than 1 in 9 emergency room visits are caused by adverse drug effects, which translates to health system costs of more than 30 billion US dollars every single year. Significant improvements in qualities of life and considerable health care savings can be realized if one could improve drug dosing and monitoring by understanding how medications are metabolized and cleared from the body. However, despite significant research efforts, scientists and clinicians still do not fully understand all of the external factors influencing drug clearance. One important factor that has been overlooked, perhaps until now, is "music". Music exposure, which is ubiquitous, is known to affect hormone (e.g., testosterone, estrogen), neurotransmitter (e.g., dopamine), and cytokine (e.g., interleukin-6) concentrations in various pre-clinical and clinical models. As these compounds are synthesized and metabolized by the same clearance pathways as drugs, we hypothesize that music can elicit differential effects on the activities of metabolism enzymes and/or transporters responsible for drug clearance. Our objective is to systematically investigate the effects of different modalities of music (i.e., tempo, rhythm, genre, harmony, dynamics, auditive frequency, and vibrational kinesthetic frequency) on select, major drug metabolism and excretory pathways. Our initial approach is to characterize the metabolism enzyme activities in rodents exposed to music using an innovative, high-throughput in vivo-in vitro experimental design (minimizing animal sacrifice). We will also translate these findings to humans by examining the effects of music on the pharmacokinetics of endogenous/natural probe substrates which are selective for specific metabolism and transporter enzymes (minimizing the invasiveness of the clinical experiment). We will utilize the principles of equity, diversity, and inclusivity, as well as sex/gender-based analyses in our experimental designs. We expect our novel paradigms, if proven successful, to be able to impact a very broad segment of the human race. Specifically, music may be effectively introduced into mainstream medicine (e.g., clinicians prescribing Mozart Concerto with patients' diabetic medications), disrupting conventional thinking. Moreover, music could also potentially reduce the incidences of adverse drug effects, leading to improved quality of health care and lowering system costs associated with hospital admissions.

Carbon dioxide (CO2) is the main greenhouse gas driving climate change. New technologies that sequester it are key to mitigating global warming and innovative and interdisciplinary carbon capture solutions are needed. One strategy involves leveraging CO2-fixing cyanobacteria. Cyanobacteria adapt to varying CO2 levels by regulating processes that facilitate its fixation. However, the molecular mechanisms by which CO2 is sensed by cyanobacteria, or any organism, remain unknown. Identifying these sensors will therefore shed light on fundamental biology and has potential high-reward by enabling engineering of cyanobacteria to enhance carbon capture.

One possible sensing mechanism involves CO2 reacting with select lysine residues and N-termini of proteins to form Protein-CO2. Such CO2 modification of proteins reversibly alters both structure and charge, making it ideal for CO2 sensing. But current methods to detect protein-CO2 are inadequate and so this modification is little explored.

Objectives: We hypothesize that protein-CO2 is a functional switch that enables organisms, including cyanobacteria, to sense and adapt to CO2 levels. The goal of this proposal is to determine how protein-CO2 enables regulation of CO2 fixation in cyanobacteria. Specifically, we will address the following objectives:

Obj.1: Discover novel protein carboxylation sites in cyanobacteria.

Obj.2: Identify CO2 sensor proteins that regulate CCM expression.

Obj.3: Characterize the role of protein carboxylation as a mechanism of CO2 sensing.

Approach: We will develop high-throughput chemoproteomic methods to identify protein-CO2 sites in cyanobacteria. In parallel, we use a genetic reporter system in conjunction with saturating transposon mutagenesis to identify genes that enable adapting CO2-fixation to CO2 levels. These datasets will enable identifying CO2 sensor proteins regulated by protein-CO2, which will then be characterized biochemically and in live cells to define their CO2 responsiveness.

Novelty and significance: Our interdisciplinary approach and new methods will enable interrogating protein-CO2 in any organism and help establish this modification as a CO2-sensing mechanism. This finding would challenge conventional thinking as CO2 sensing is generally thought to be indirect, occurring through its hydrated form (HCO3-). Knowledge of the CO2 sensing mechanism of cyanobacteria can be leveraged by bioengineers to develop improved CO2 capture technologies.

OBJECTIVE:  We will create a novel  technology for non-destructive inspection of fiber orientation distribution (FOD) in discontinuous, long fiber reinforced polymer (DLFRP) composites. The objective is to develop an inspection technology that can assist in quality assurance of the FOD in compression molded DLFRP composite parts, and overcome the limitations of the existing, conventional imagery-based inspection methods.

HIGH RISK: Due to the weight and economic savings they provide, compression molded DLFRP composites gained importance in the aerospace and automotive industry, in which Canada has a significant presence and strategic interests. Quality assurance of the FOD is a grand challenge in advanced manufacturing of DLFRP composites - this is because high-throughput manufacturing methods, such as compression molding, typically result in complex, locally varying, stochastic FOD, which translates into part-to-part variability and, consequently, uncertainty in mechanical performance characteristics of produced parts. Poor production quality control results in significant operational and financial costs as high as 15-20% of annual sales revenues. Replacing metal parts with compression-molded LDFRP composite parts results in up to 40% weight savings which translates to energy savings, fuel economy and reduced global emissions.  Many industries are replacing metal parts with composite parts, but the challenges involved in controlling the FOD quality and repeatability of the mechanical properties have limited this to only about 5% of the potential replacement.

INTERDISCIPLINARITY:  This proposal requires an interdisciplinary approach to FOD inspection based on the combination of the state-of-the-art composite manufacturing process modeling, traditional thermal imaging, and advances in machine learning, Bayesian inference, and computer model emulation and uncertainty quantification. Traditional imagery inspection methods (such as optical microscopy and volumetric X-ray micro-computed tomography) are often time consuming, destructive in their nature and/or suffer limitations on the physical size of the specimen which can be analyzed.

HIGH REWARD: The outcome of this project will be a technology for FOD inspection that would help Canadian companies to control the quality and repeatability of the mechanical properties of compression-molded DLFRP composite parts.

Evidence of life on early Earth (3.45 Ga) comes from cyanobacteria that formed conical stromatolites. Based on the fossil records and laboratory experiments, it is currently believed that the conical structures developed in shallow waters (<20m) and that their cm-scale spacing was a physiological response to reduce competition for nutrients. We propose to challenge this paradigm which, if proven, would provide new insights in the evolution of life on early Earth.

Perennially ice-covered lake Untersee (Antarctica) is the only known lake with actively growing large conical stromatolites that are similar in morphology to those in the fossil records. The stromatolites are found to depths of 160m and spaced at the m-scale, which suggests that the development of conical structures is not limited by hydrostatic pressure and bubble formation (<20m depth) and their spacing might not be associated with nutrient availability. The project will quantify the spatial organization of the stromatolites and link their distribution to physiology and competition for limited nutrients and light. With the acquisition of georeferenced high-resolution imagery (by a submersible remote operated vehicle and divers with scuba) combined with limnological measurements and sampling of the stromatolites, the objectives are to:

1) Determine why Untersee developed cuspate and conical stromatolites, while other lakes didn't, by mapping their depth, size, orientation and light availability across the lake. This will answer if the morphological diversity and unexpected m-scale spacing of cones is part of an ecological response to the persistent ice cover, low-light, CNP-starved ecosystem.

2) Identify metabolic adaptations that enables the survival of microbes in conditions near the limit of life by identifying community structure and function and conducting experiments on photosynthetic prokaryotes.

3) Determine if the water column and sedimentary biogeochemical cycling plays a constitutional role in the formation of the pinnacles and cones by using stable isotopes measurements and tracers.

Our high risk study, combining several sciences with cinematography, will provide novel insights into an ecological enigma: the evolution of early life. Further, our activities will advance ongoing efforts to designate Untersee Oasis as an "Antarctic Specially Managed Area " and contribute to Canada's mandate to support research in Antarctica to join the Antarctic Treaty as a full voting member.

This interdisciplinary and intersectoral research proposal aims to provide remote communities with energy storage systems that do not rely on fossil fuels. The electrification of remote communities is a global issue, part of the 17 UN Sustainable Development Goals. There are nearly 200,000 people in Canada who are not connected to the electrical grid. These communities must rely on diesel generators or solar and wind systems combined with lead or Li batteries. Such systems are polluting and/or use expensive, non-renewable materials, leaving these communities vulnerable to supply disruptions. Some of these remote, often indigenous, communities also have a complicated history with large-scale infrastructure projects. Electricity storage systems with biobased organic redox flow batteries (RFB), coupled with renewable energy, are an interesting alternative solution as a path to carbon neutrality, allowing the use of intermittent wind and solar. Instead of using electrochemically active (redox) compounds in the solid-state as in conventional batteries, RFBs use redox molecules in solution, thus in the liquid-state to store electrical energy. Based on aqueous solutions, RFBs are safer (non-flammable) and ideally suited for large-scale storage, with possible electrolyte replacement. RFBs' electrolytes are currently based on vanadium, a non-renewable element. Yet, to be economically viable and environmentally sound, RFB redox electrolytes must be produced sustainably. This project aims to develop biosourced organic molecules that will perform as well as current electrolytes. 3D printing will be used to quickly generate different flow cell architectures, adapted to the new molecules' cell requirement to maximize their performance. How technological devices create social values (equity, inclusion, social cohesion, social utility) is not well understood and not often studied. To that end, we propose to examine the disposal of the RFBs' components at their end-of-life to optimize the natural resources used to make them. We hope to integrate the social values of inclusion and equity in the very design of the device. We also aim to cross the original boundaries of design and technology adoption to qualify and quantify the social utility of these RFB batteries as to their economic potential. We will account for the organizational constraints that could legitimize or, on the contrary complicate, the societal scope of our initiative.

Background: Chronic wounds cause significant socioeconomic burdens to our society and remain a leading cause of lower limb amputation (6.5 million patients in the US). Impaired healing of chronic wounds is multifactorial, but infection of various pathogens, poor vascularization, and excessive pro-inflammatory factors (e.g., cytokines and reactive oxygen species) are amongst the most commonly found players. Treatments for chronic and infected wounds remain an unmet clinical challenge, calling for multidisciplinary efforts to invent novel biomaterials to transform chronic wound care.

Paradigm shifting: Current repertoire of wound care materials are all polymeric materials without/with therapeutic payloads. They are largely mechanically passive and non-living. The existing paradigm avoids bacteria due to biosafety concerns. In contrast, our new paradigm capitalizes genetically-modified non-pathogenic bacteria as building blocks to form a new class of engineered living materials. The resulting living bacterial bioadhesives could enable an unprecedented combination of living mechanics and programmable functionalities for chronic wound care.

Objectives: We hypothesize that genetically and chemically engineered non-pathological bacteria could form living functional wound dressings, capable of protecting and healing chronic wounds while battling with infecting bacteria. Specific Aims include: (1) synthesis and characterization of the living bacterial bioadhesives; (2) genetically modifying the bacteria within the bioadhesive to sense and destroy multidrug-resistant pathogens; (3) programming them to release potent agents to promote wound healing. 

Approaches: This project draws upon expertise of leading researchers in biomaterials, biofilm engineering, wound healing, and microbiology. Multidisciplinary approaches include surface and genetic engineering of bacteria, biorthogonal click chemistry, in vitro culture, ex vivo human skin study, and in vivo animal models using clinically relevant pathogenic bacteria.  

Novelty & significance: This application could advance science and technologies in biomaterials, biofilms and wound management. The paradigm of engineered living materials is innovative and transformative. Using genetically modified bacteria to battle infection is paradigm-shifting. This high-reward project is linked with ever-increasing market of wound care ($8.5 billion by 2021) and benefit millions of patients living with chronic wounds. 

Disease diagnosis often involves complex samples such as blood, urine and saliva, or complex targets, such as cancer cells and exosomes. The typical method of diagnosis is to measure the concentration of a particular compound such as glucose or a particular cell surface protein using a specific recognition element such as an antibody or an aptamer. While this method works well for diseases like diabetes, for diseases such as cancer, neurodegenerative diseases and immune diseases, finding a single biomarker is difficult, since many related molecules might change at the same time. To solve this problem, having a sensor array to form a recognition pattern is an alternative approach. While many sensor arrays have been reported, the number of sensors in the arrays is often quite limited (e.g. less than 10). Often times, the exact target molecules are unknown and researchers have to use a trial-and-error approach to design the sensor components in an array.

Aptamers are short single-stranded DNA that can selectively bind to target molecules. Aptamers are often selected via multiple rounds of selection with the goal of obtaining the best binding sequence for target molecules. We hypothesize that a complex sample needs to be analyzed using a complex sensor array. Instead of getting a few specific binding sequences, we aim to obtain sequence patterns to match for disease states.

Using a highly interdisciplinary approach, we aim to combine single-round DNA aptamer selection with artificial intelligence (AI) based pattern recognition to achieve this goal. Our aptamer library will contain billions of unique DNA sequences, much larger than any other sensor array. Each sequence can serve as a sensor element. After incubation with the target sample, the binding DNA will be sequenced using a commercial nanopore sequencer and the sequence patterns will be analyzed using an AI algorithm, which will take into consideration the primary sequence, secondary structure, and sequence abundance of the final library.

Our initial tests will be performed on a few cell lines including liver cancer and normal liver tissues. It has a high risk due to the complexity of both the sample and the sensor array, but the maximal information content in the array will make it a high reward project if successful and it provides a general yet powerful solution to disease diagnosis.

Stroke, as the second leading cause of death worldwide, significantly impairs the activities of daily living for patients. Its impacts upon many aspects of a patient's life require targeted rehabilitation to address the personalized sequela. The current clinical practices usually need ongoing and complex assessments in a hospital or clinic, leading to inconvenience and inconsistency of gains for patients. Thus, an "always-available" assistance for stroke patients is of paramount importance. With recent advances in wearable electronics, it is foreseeable to realize a "rehab-clinic-on-skin" where all the current assessment (e.g., motion measurement and electromyography (EMG)) and treatment (e.g., functional electrical stimulation (FES)) tools are integrated onto a patient's skin imperceptibly. To achieve that, stretchable electronics mechanically matching the human skin need to be developed. Currently, there is still large gaps on conductive materials and stretchable electronics design.

To tackle these challenges, we will develop an ultrasoft and inherently adhesive bottlebrush elastomer (BBE) mimicking mechanical properties of different tissues such as skin. With the Young's modulus orders of magnitude lower than those of commonly used elastomers, a BBE substrate provides much better conformity with human skin and has imperceptible physical effects to patients. We will also develop a liquid metal/silver composite that can be patterned onto the ultrasoft BBE with extraordinary conductivity, stretchability, and adhesion, which will be used to construct BBE-based printed circuit boards (PCBs). We will integrate inertial measurement (IMU) modules, EMG/FES modules, and wireless communication modules into a BBE-based PCB to achieve ultrasoft, stretchable, and adhesive electronics on skin for upper-limb rehabilitation.    

Stroke imposes heavy burdens on the quality of life for patients and their families. To improve the comfort and provide "always-available" rehabilitation for patients, we design new material platforms and "clinic-on-skin" devices, which are highly demanded for both patients and hospitals or clinics. Our study combines intradisciplinary research of materials science, mechanical engineering, electrical engineering, and occupational science. We propose a new concept of "rehab-clinic-on-skin" for helping patients and improving their quality of life.

Bacterial resistance threatens the clinical use of all antibiotics. Beta-lactams (e.g., penicillins) are the most widely used antibiotics, yet resistant bacteria readily degrade them by producing beta-lactamase enzymes. Clinical detection of beta-lactamase producing bacteria is routinely carried out with phenotypic assays. However, these tests require bacterial culturing, during which time bacterial infections can worsen and resistant bacteria can spread.

We propose to develop a new diagnostic device that can rapidly detect beta-lactamase-producing bacteria directly in patient samples. To accomplish this, we will apply surface-enhanced Raman spectroscopy (SERS), an extremely sensitive technique that can profile the composition of complex samples. Raman spectra contain distinct bands corresponding to different chemical functional groups, and beta-lactams manifest with a pattern of characteristic bands.

Our objectives are to: (1) Develop a new, ultrasensitive, SERS-based "fingerprinting" method that  permits rapid detection of clinically used beta-lactams and their hydrolysis products. (2) Integrate this method into a SERS-microfluidic platform capable of directly testing clinical samples (e.g. urine, serum). (3) Explore the potential of this technology to serve as a unique tool to monitor the degradation kinetics of antibiotics (e.g., beta-lactams) by antibiotic-resistant bacteria. As a first step, we will study the degradation of beta-lactams with purified beta-lactamase enzymes and pure cultures of resistant bacteria. Beta-lactamase-producing bacteria in these samples will be detected by treating the sample with an antibiotic and using SERS to monitor antibiotic depletion and/or the formation of degradation products.

Our interdisciplinary team of Raman experts, microfluidicists, microbiologists, and biochemists is strategically positioned to develop and validate this technology. Unlike other assay platforms, our device will detect resistant bacteria within an hour from patient sample acquisition. In addition, this technology will personalize the treatment of bacterial infections. By screening a patient sample using a panel of beta-lactams, antibiotics not degraded by resistance enzymes can be identified, allowing clinicians to prescribe the optimal antibiotic treatment for a particular bacterial infection. Our proposed technology can also be applied to monitor the degradation and/or metabolism of other drugs directly in patient samples.

Governments are facing increasing pressure to address complex challenges like climate change, growing inequity, reconciliation with Indigenous Peoples, affordability, and many others at the scale and rate that these challenges demand. Innovation in the public sector is quite urgently and desperately needed, and governments must begin to adapt the paradigms, processes, systems, structures, and tools of their trade to better serve those they are responsible and accountable to. But how might an organization best known for its reliability, consistency, and slow rate of change become more innovative, adaptive, and responsive?

There is a global proliferation of Public Sector Innovation Labs (PSI labs) responding to these pressures, with estimates that more than 500 now exist in all parts of the world. Many network-serving organizations are emerging and shifting their focus to support PSI in the various ways that it is defined and practiced, yet most are focused on finding efficiencies and improvements through technology, user-centered design, and cost effectiveness. There is unexplored potential to learn with/from public sector innovators that hold an ambitious and transformative intent in their work. This is a field-in-motion that is ripe for applied research and knowledge mobilization.

Of particular interest in this course of study are urban and city-oriented PSI labs, capacity building initiatives, and related activities/interventions that hold a transformative, emergent, and/or resurgent innovation intent, and make a more distinct shift away from dominant governance regimes and toward imagining and enacting other possible futures. This research centers on public sector innovators working on changing the systems, structures, and paradigms of urban governments to get to dramatically indifferent outcomes that move toward social and ecological justice for all.

This action research will initiate, grow, support and enable a global community of practice of PSI labs and network-serving organizations leading in transformative innovation and approaches to tackle these complex challenges, like implementing the United Nations Declaration on the Rights of Indigenous Peoples in cities. The purpose of this action and applied research is to support experimentation, learning, and higher impact interventions and outcomes of these public sector innovators, and to mobilize this thinking, learning, and knowledge with relevant research and practitioner communities.

Immune cells secrete cytokines to communicate with each other in order to coordinate the intricate processes that allow the body to eliminate pathogens, repair damaged tissue, and suppress cancer. Dysregulation in cytokine secretion by one or more groups of immune cells may disrupt immune homeostasis and trigger pathologies such as sepsis, cytokine storm (e.g. from severe COVID-19), and chronic autoimmune diseases. Consequently, there has been a long-standing challenge to couple specific immune cells with their secreted cytokines in order to untangle the complex cytokine signalling that drives pathological processes.

Immune profiling currently involves characterizing (1) the cytokine secretome in serum (e.g. using ELISA or Luminex beads) and (2) immune cell composition in cellular samples (e.g. using flow cytometry or single cell RNA sequencing). These two types of profiling data are currently disconnected, which obscures the context of cytokine signalling. In effect, this approach muddles the connection between the messages and the messengers of the immune system. Current methods attempt to address this issue using intracellular cytokine staining or transcription of cytokine genes, but these surrogates are often not correlated with cellular secretion.

This project will develop a new technology to simultaneously profile cytokine secretome and transcriptome on single cells in order to establish the linkage between the cytokine signals and their source cells. This work will demand perspectives and expertise from engineering, chemistry, genomics, bioinformatics and immunology in order to create a unique technology to simultaneously capture the functional cytokine secretion profile alongside the nuanced phenotyping available through single cell RNA sequencing. This technology will give researchers a new tool for understanding how immune homeostasis is maintained through cytokine signalling, as well as how such states are disrupted during disease. More broadly, this capability will allow researchers to challenge existing paradigms of diseases heterogeneity and subtyping in order to classify diseases based on immunological status to develop appropriately tailored treatments to individual patients.

Over the last 20 years, the increased use of non-invasive ventilation has enabled millions of children, from infants through teenagers, and adults to use breathing support at home rather than needing to remain in hospitals or healthcare facilities. While this increase in the accepted use of home-based non-invasive ventilation has been life changing for non-invasive ventilation users and their families, there has been relatively little advancement in the mask interfaces used to deliver this life enhancing and sustaining therapy. Standard, mass produced commercial masks provide reasonable fit for the average face but are often a poor fit for those of non-Caucasian race or persons with unique facial features, including growing children and those with craniofacial abnormalities.

The objective of this project is to develop a rapid and accessible process workflow to provide custom masks as part of an integrated system that helps maximize mask use for people who need them. We aim to reduce barriers and improve access so that those in remote communities have equal access to custom masks. We will leverage the latest developments in 3D facial scanning, artificial intelligence-computer aided design, additive manufacturing, and sensor technology.

We will focus on children and adolescents given the unique challenges that growing faces and changing interests and behaviour present. Persons with lived experience with non-invasive ventilation will be integrated into our multi-disciplinary team of scientists and clinicians as equal members to ensure our process if informed by those with the expertise in using this technology. Customization of the mask shape will be driven by 3D imaging of the face, exploring options such as smart phones to capture facial images to reduce cost and need to travel to access a specialized 3D imaging system. We will explore silicone printing and thermoplastic molding to reduce mask production time. We will integrate feedback on fit and air leak into the mask design to provide the end-user with information to improve comfort and ease of use. To make the masks more attractive to end-users, we will explore options for individual customization with colour, personalized patterns, or other means of making the mask identifiable as their own. Collectively, these initiative will support an integrated, end-user-centric process of mask customization, performance and adherence monitoring, and mask replacement to respond to changing needs.

The brain integrates sensory stimuli into low-dimensional representations that aid in cognitive functions such as navigation, planning and memory. Specifically, the firing activity of neurons of the hippocampal formation forms a `cognitive map' that can represent both spatial location and task-relevant latent states. The global structure of the cognitive map is the brain's coordinate system for performing a task. Parametric, interpretable, unsupervised modeling of this structure is key to unlocking the algorithms underlying cognition and complex behaviour. We will tackle this challenge via electrophysiological recordings in behaving rats, analyzed using novel nonlinear manifold learning techniques that leverage deep generative modeling, and topological, algebraic and geometric foundations of group theory.

Approach:

(1)   We will perform neural recordings in the hippocampus and prefrontal cortex of rats in a virtual reality (VR) apparatus where they perform abstract, non-spatial tasks with prescribed underlying topologies and symmetries. (e.g. choosing between conflicting reference frames informed by VR cues based on reward availability)

(2)   We will develop group variational autoencoders (G-VAE), deep generative models that identify symmetries in neural representations. A G-VAE enforces group symmetries in its latent space to fit the topology and potentially highly nonlinear geometry of the cognitive map present in the experimental data.

(3)   We will learn the latent state of the cognitive map using G-VAEs, and using neurally closed-loop experiments, evaluate how task structure is represented during learning and manipulations of sensory input and task specification.

Our research plan tackles two grand challenges. In neuroscience, we seek to quantify the flexibility of the cognitive map: how the brain integrates sensory and cognitive information into a dynamic neural tapestry carefully tailored to the task. In machine learning, we seek to build architectures that learn and conform to symmetries in the data, in comparison to current techniques that can only take advantage of known structures in the data. Our research plan is highly ambitious - we are an interdisciplinary team of ECRs developing techniques transformative to fundamental neuroscience and big data analysis.

Tympanostomy is the most common surgery performed on children in Canada and the US, and is the current standard clinical treatment of chronic and recurring otitis media. Otitis media (infection and/or accumulation of liquid in the middle-ear space) is a common childhood disease and the second most prevalent cause of hearing loss and the leading cause of health care visits and antibiotic prescriptions. Its global burden is fifth among all diseases, affecting 1.23 billion people worldwide. Canada is especially affected by this disease, as its prevalence rate among Inuit, First Nations and Métis children in some Northern Canada communities is as much as 40 times that found in US and Canadian cities. In children, otitis media can have significant consequences leading to speech and language development disorders, educational problems, physical suffering, emotional distress and behavioural problems.

Tympanostomy surgery involves making a slit in the tympanic membrane, and placing a ventilation tube inside the slit. To successfully perform tympanostomy, the patient's head must be held stationary, requiring general anesthesia in children. Tympanostomy has other drawbacks such as long-term damage to the tympanic membrane and economic burden due to surgery and other medical and indirect costs.

In this project, we will develop a paradigm-shifting device for low-risk treatment of otitis media. The device is a patch that includes a collection of hollow microneedles that provide enough ventilation to treat otitis media while minimizing the long-term damage to the tympanic membrane. We will also develop a medical gun that can be placed in the ear canal to implant the patch during an office visit without touching the tympanic membrane. The combination of the patch and its gun will address all drawbacks of tympanostomy including the need for general anesthesia.

This high-risk high-reward project is only possible using a novel interdisciplinary approach that interweaves several disciplines from biology to ballistics to solve a problem that spans from human health to early education. Unlike recent efforts to develop new tools to automatize the tympanstomy procedure, this project completely challenges that extensively established paradigm. It will make low-risk and low-cost treatment of otitis media accessible not only to Canadian children, including those in Northern communities, but also to children around the world and will improve their quality of life.

Canada is among the few seismically prone countries that did not suffer earthquake casualties in the last century. However, this resulted in a very low-risk perception, inadequate preparedness and absence of data on how people would respond to the next destructive earthquake, creating the ideal conditions for a catastrophic event. The proposed research will challenge established paradigms by unprecedently bringing together domain knowledge in anthropology, earthquake engineering and computer science, to devise an interdisciplinary framework for exploring - in virtual dimensions - the response of diverse Canadians to simulated seismic disaster scenarios.

Multiple cutting-edge contributions will be offered to address the following grand challenges: i) enabling district-scale reliable simulations of building collapses, not feasible via standard structural analysis, ii) creating videogame-inspired virtual reality environments duly portraying post-earthquake scenarios, outperforming previous static animation models, iii) observing behavioural reactions of real-world individuals exposed to virtual seismic hazard, critical data not collectable in actual earthquake aftermaths. In phase I of the study, digital twins of selected high-risk urban districts will be created, alongside a mapping of perceived risk by local occupants. In phase II, the unexpressed potential of physics-based engines, presently confined to computer graphics, will be exploited by adapting them to the real-time mechanics-based prediction of structural failures. In phase III, a sub-set of local occupants surveyed will be invited to participate in immersive earthquake gaming experiences for investigating live reactions, relationships with perceived risk and environment, and further tailor the framework to user's experience. Through this study, we will gain a user-informed understanding of social and structural impacts of earthquake hazards, while providing new tools to assess and mitigate associated risk.

This project will lay the foundation for the next generation of holistic seismic risk assessment tools, enabling us to uncover the complex interrelations among human behaviour and earthquake hazard. The devised digital framework, readily adaptable to simulate different types of calamities, will also improve disaster preparedness by providing decision-makers with a first-of-its-kind platform for virtual search and rescue training, evaluating seismic losses, improving emergency management.

The COVID-19 pandemic has left the world reeling. As we begin to recover from the initial crisis, we must begin to address the long-term impacts. Among these are high numbers of COVID survivors displaying "long COVID" with persistent symptoms including recurring headaches and "brain fog." As of now there is no clear clinical consensus on the cause or even the symptomatic presentation of long COVID, much less a measurable biomarker or understanding of the mechanisms behind it. Sufferers of long COVID, like other invisible disabilities, may be subjected to disbelief in the absence of an easily provable diagnosis. We propose to use medical imaging, computational modeling, science journalism, and disability studies techniques to produce a unique multimedia experience from which we will gain understanding of the possible underlying mechanisms and engage with and disseminate the experiences of scientists, research participants, and the general public in the context of a scientific study.

The twin primary objectives of the research proposal are:

(i) training of a generative deep learning model on a series of quantitative medical images of the brain to produce a physiological connectome, representing the ways in which different imaging properties are connected within the brain;

(ii) use of scientific journalism practices and disability studies techniques to effectively document the experiences of selected participants and scientists.

We will perform a series of brain scans of a patient cohort that do not report long COVID and a patient cohort that do report long COVID and will employ the resulting data to train two deep learning models, one representing a healthy brain and one a brain affected by long COVID, which we will analyze using network-based techniques. These models will help devise biomarkers of long COVID in the brain, a first step towards developing treatment strategies. Simultaneously, through a series of interviews, we will document the process and experience of a subset of people involved in the study and derivation of a generative model of the brain, getting input from principal investigators, graduate researchers, and study participants who have and have not experienced long COVID. By presenting it in an experiential fashion, we hope to challenge the traditional divisions between scientists, participants, and public and demonstrate clearly (whether or not we are able to find a truly measurable outcome) that the submeasurable is not unreal.

A keystone feature of the Arctic is the snow cover that develops over the long winter. This snow cover plays a role in controlling climate, hydrology, lake ice cover, permafrost, greenhouse gas fluxes, and ecosystem function for example. However, measuring Arctic snowfall and resulting snow on the ground is prone to extremely large errors, with measured snowfall typically much less than actual and less than measured snow on the ground. To solve the challenge of accurately measuring snowfall, requires addressing the issue of snow gauge undercatch during periods of high winds. Although known for decades, this undercatch problem has been shown to be surprisingly complex. As a result, published snowfall data, typically significantly underrepresents actual snowfall and it is left to the user to "correct" these data. This is a significant problem as these data are relied upon by climate, weather, and water agencies, and researchers in order to document changes in the Arctic snowscape, predict streamflow for hydroelectric generation, and to test and validate water and climate predictive models. Our team is proposing a unique interdisciplinary study that will use recent advances in applying upward looking weather radar, eddy covariance, cosmic ray sensors, unmanned areal systems, and next generation snow models to develop new snowfall measurement methods. The measurement and modelling of Arctic snowfall is high-risk research for numerous reasons. First, despite considerable expertise from team members and access to an Institutional Arctic research station, there is significant risk associated with maintaining a complex array of instrumentation over the Arctic winter. Second, integrating these data with a novel ultrahigh resolution snow model that has limited Arctic testing is high-risk. However, our ongoing research suggests this approach may result in dramatic advances and, if successful, will provide much needed methodology to accurately measure Arctic snowfall for the first time. In this proposal we outline the integration of three interdisciplinary research areas: 1) novel snowfall systems and atmospheric sounding methods to better understand and measure snowfall, 2) new methods to measure average snow water equivalent across a small watershed, and 3) implementation of an ultrahigh resolution snow model to integrate and compare #1 and #2. The A-Snow team will integrate these interdisciplinary components to provide a novel method to measure Arctic snowfall. 

Liver disease is the tenth most common cause of death in Canada. The diagnosis of liver disease is based on non-specific biomarkers of liver injury and invasive biopsies for histological assessment. However, no test to define liver function exists which has disallowed new advances in the field of liver disease and hindered the development of novel liver therapeutics. Here, we plan to develop a new quantitative modality for testing liver function based on liver-targeted polymersomes. This approach is novel since these polymersomes will specifically target hepatocytes contrary to untargeted nanoparticles traditionally used which are engulfed by the dominant Kupffer cells. This will revolutionize outcomes for clinical trials in liver disease and change management since function is a precise qualitative evaluation compared to routine tests used to date.

Engineering strategy. We propose to develop fluorescent hepatocyte-targeting polymersomes and validate them in various rodent models of liver disease. The polymersomes will be designed to preferentially accumulate in hepatocytes due to 1) their small size (80 nm) which enables them to pass through the fenestrations of the liver endothelium to reach hepatocytes; 2) a covalent surface modification with the established hepatocyte-targeting agent for binding to hepatocytes; 3) their high polyethylene glycol content of 100% which makes them "stealthier" and thus less prone to Kupffer cell uptake than conventional nanoparticles.

Driving hypothesis. Uptake of hepatocyte-targeted polymersomes defines hepatocyte function.

Aims

1) Synthesis of fluorescent, biodegradable, and functionalized block copolymers and polymersome preparation

2) In vitro specificity testing (primary hepatocytes and Kupffer cells)

3) Validation in different rat models of liver disease

Novelty and significance. As traditional nanoparticle formulations show high Kupffer cell uptake, we are engineering a system with unique attributes to specifically target uptake in functional hepatocytes and not in Kupffer cells. Our quantitative hepatocyte-targeted test will reflect the functional status of the liver and thus transform the diagnosis of liver disease, enabling earlier diagnoses, better disease stratification, and assessment of treatment response in clinical trials and practice. After the fluorescence-based proof-of-concept, the polymersomes can easily be tagged with a positron emission tomography marker for translation to clinical practice.

Most Alzheimer's disease research completed to date has been directed at understanding the rare familial version (10%). Little is known about the etiology of the more common (90%) sporadic form of Alzheimer's disease (SAD). One view suggests that this brain disease manifests itself by the presence of various combinations of co-factors. An individual SAD patient would have some combination of these factors, but the type and combinations could be different from another patient with similar clinical symptoms.

This research is based on recent studies showing that approximately 100 gene polymorphisms are associated with SAD and that they cluster into functional groupings. The objective of the proposed research is to test the intriguing idea that gene polymorphisms increase susceptibility to SAD by making the brain more vulnerable to the presence of other co-factors during adulthood. The gene polymorphisms of interest for this program were selected to represent this heterogeneous clustering, including mutant mice with alterations in the following genes: 1) TREM2; and 2) ECHDC3. The gene mutations are of interest for their potential roles in inflammation (TREM2), and insulin regulation (ECHDC3). The impacts of various modifiable co-factors in these genetic predisposition models of SAD will be assessed including stress, or unhealthy diet. Next, we will test the impacts of different cannabinoid extracts that will be formulated to target the specific functional pathways thought to be altered by these gene mutations. The cannabinoids are of great interest as a potential treatment for neurodegenerative diseases because they are currently available, relatively safe, and target multiple pathways implicated in these diseases.

Project 1: novel mutant mice models of SAD alone or in combination with different lifestyle modifications (stress or diet) will be assessed for brain and body changes such as: metabolomics; proteomics; plasticity mechanisms; cholinergic status; and memory.

Project 2: will assess the impacts of targeted cannabinoid treatments on mutant mice with different lifestyle modifications on brain pathology, body changes, and cognitive impairments associated with SAD.

Novelty and significance of the proposed research: 1) develop novel rodent models of SAD, 2) explore treatments that target specific mechanisms implicated in the disease state specific to each SAD subtype; 3) clear translational implications for human SAD.

This project challenges and complicates the dominant paradigm in surveillance studies that approaches surveillance in terms of power and coercion, considering it instead through the lens of pleasure. While most surveillance research has focused on the harmful or coercive, few have approached it in terms of pleasure. This project conceptualizes surveillance not only as a tool/technology but a methodology, ideology, and way of being in the world. It asks: is pleasure a generative way to use surveillance, especially for marginalized and surveilled groups? Can surveillance bring pleasure through empowerment, connection, self-surveillance, exhibitionism/voyeurism, play, or art? Using collaborative research-creation methodologies, the project will (a) engage artists from BIPOC, 2SLGBTQIA+ (queer), disabled, and other marginalized communities to create or contribute artistic work exploring these questions, culminating in an art exhibition and catalogue; (b) produce a zine of pleasurable surveillance strategies; and (c) produce a monograph exploring questions of pleasure and surveillance. The risk of this project is that it disrupts the current dominant paradigm in surveillance studies by bringing it together with research-creation, two very different areas. Surveillance studies is a transdisciplinary field that primarily draws influence from disciplines such as policy, criminology, sociology, law, and communication studies, whereas research-creation is a methodology rooted in creative and artistic interventions (Manning, 2016; Loveless, 2016). There is always risk in engaging marginalized peoples. The high rewards of this project include (a) a new theoretical-methodological paradigm for surveillance studies that enriches dominant approaches and suggests new directions for research; (b) new strategies for marginalized communities to use surveillance in pleasurable ways, and (c) demonstrates generative uses of research-creation in surveillance studies. While artists and scholars have engaged with surveillance art, there is limited research drawing specific influence from research-creation as a methodology. Using a novel methodology challenges surveillance studies because it not only validates creative interventions as knowledge producers, but it disrupts traditional ways of researching in the field of surveillance studies. We propose to adapt research-creation as a methodology to rethink and revisit the traditional theories embedded in surveillance scholarship.

Every year, close to a quarter of a million of anterior cruciate ligament (ACL) injuries occur in Canada and in the United States. Being one of the key ligaments that help stabilize the knee joint, ACL repair is critical. ACL tears do not heal without treatment and reconstruction surgery remains the major orthopaedic procedures where an orthopaedic surgeon entirely discards the torn ACL and reconstructs the ACL with an autograft. This procedure is generally associated with the morbidity of a graft harvest. Reconstruction surgery with autograft or synthetic grafts has been associated with a high rate of failure due to a lack of vascularity, loss of blood clot as well as important fluid movement inside the joint causing bioactive molecules to be washed out and preventing cell adhesion to occur. the BEAR implant, a foam positioned and sutured on one side to ligament has been developed. Unfortunately, it is not always successful to protect the ligament during healing.

To address this challenge, we are herein proposing a paradigm shift from the current standard of care. Our objective is to combine repair of the ligament with an electrospun fibrous sponge that can fill the synovial cavity and protect the ACL wound site. To this end, we design and develop an advanced technology combined with highly functional, and adhesive synthetic ligament capable of 1) simply binding the ends of the torn ligament while 2) producing a robust foam network to trap blood and stabilize bioactive compounds in the gap between the torn ligament extremities.  This "biomaterials-by-design" concept with fluid flow simulation and the resulting grafts will further be validated via in vitro and in vivo pilot tests.

Reflecting the advanced nature of the materials to be developed, processed, and assessed, this high-risk and high-reward NFRF project can only be achieved by using a transdisciplinary team of researchers during every stage. Upon completion, we anticipate producing advanced biomaterials and achieving innovative processes and unique approach to tackle failures in soft tissue repair and regeneration. This is going to become the first biomedical technology capable of healing or restoration of torn ligament or tendon since the standard care procedure initiated more than 30 years ago. This will help all individuals suffering from damaged soft tissues resulting in chronic pain with a significant impact on quality of life, productivity and accumulated recourse to the health system

This high-risk, high-reward multi-disciplinary tri-council project aims to develop a hallmark precision critical care solution to enable personalized management of over 200,000 admissions annually to Canadian Intensive Care Units (ICUs). Patients with acute illnesses and injuries are cared for using advanced systems for organ support but ~1 in 10 of patients will not survive their stay. While clinicians at the bedside can identify sustained critical events, fleeting episodes may go unnoticed without potentially life-saving early interventions. In the face of the recent pandemic, it has become clear that this inadequacy demands new solutions so that the right patients are treated with the right therapies, at the right time.

Bedside monitoring is a staple of ICU resulting in large amounts of physiological signals. AI-enabled methods have been reported for automated detection of important events in ICU, however, AI of today is not able to adequately address the challenges of precision critical care. ICU data are highly noisy due to sensor drop-off and motion artifacts; added to this is the lack of ground truth labeling for the massive amount of data, and the uncertainty in the limited labeled data. Additionally, the significant diversity of patients and illnesses leads to insufficient representation of classes of critical events. Hence, despite achieving state-of-the-art performance in other domains, to-date, AI-enabled systems are not clinically deployed to augment clinical decision support in the ICU.

We propose to leverage large scale physiologic data in ICU and identify important clinical events and suggest personalized therapeutic strategies using methods with actionable and explainable outputs. In particular, we focus on learning from no or "weak" ground truth labels and providing interpretable decision support recommendations to human reviewers. Our solutions mimic physicians' decision making to say "I don't know" when faced with ambiguous data, and to provide a measure of confidence and be explainable in other cases. A key element is integration of human factors for decision making, and health economics in the interpretation and actioning of AI outputs.

The proposed project directly addresses NSE challenges in AI to impact health research in critical care, taking into account innovations in social sciences. The challenges associated with ICU data are shared across other domains increasing the potential impact and reward of the proposed innovations.

We propose to develop and explore a paradigm-shifting approach to the history of 20th century urbanization. Combining innovative interdisciplinary research methods from Social History and Geomatics Engineering, we propose to push the boundaries of Historical Geographic Information System(s) (H-GIS) and extend into an exciting new area: Historical Geospatial Data Science (H-GDS).

Recently, H-GIS have seen enthusiastic uptake by historians seeking to visualize and analyze trends and events, often relying on commercial GIS software. Yet, vast repositories of spatial data, such as the National Aerial Photographic Library (collected since the 1920s), cannot be effectively and efficiently used with existing GIS technologies. This limits researchers' ability to use sources like older aerial photos to investigate trends in urban growth and suburban expansion, in turn, limiting our understanding of social-economic conditions and government policies related to development. Our new H-GDS methodology proposes to unlock such underutilized repositories by developing new AI technologies. These will assemble and extract new information and development patterns. Using advanced Artificial Intelligence (AI), Machine Learning (ML) and Deep Learning (DL) technologies, we intend to (1) push the limit of AI, ML and DL to extract more and better urban land use information from historical aerial photos, and (2) develop AI-supported spatial-temporal analytical algorithms to analyze a wider array of land use information. We will select and interpret this spatial information guided by relevant sources in municipal and provincial archives.

New Brunswick presents an ideal case for studying equity in (sub)urban development. Three cities of proximate size (Fredericton, Moncton and Saint John) have established Anglophone, Francophone and urban Indigenous populations, as well as small but significant immigrant groups. All were affected by state projects of 20th century modernization. We seek to reveal how different groups lived through such changes, with a special interest in better understanding the social, economic and political goals of the famous New Brunswick 1960s Program for Equal Opportunity. The project will also result in a century-long time series of urban aerial data. Published online, this new resource will allow other researchers to interpret newly visible historical trends in modern development, and inform future municipal research and urban planning across Canada. 

Converting agricultural residues to fuels/chemicals is not yet techno-economically feasible due to low and highly fluctuating costs of their fossil derived counterparts, high transportation cost, and sophisticated/harsh conversion methods that cannot be decentralized. Moreover, it does not provide any benefit to the agricultural community. Hence, we aim to develop a novel strategy to make this a reality, technologically and socio-economically. The challenge is to develop a method to reduce the recalcitrance of lignocellulose for its "at source" fractionation into highly valuable cellulose, hemicellulose/sugars, and lignin. Biomass recalcitrance is a result of the lignin carbohydrate complex (LCC) linkages present in it. Hence, instead of directing the efforts purely on developing mechano/thermo/chemical methods to break down these linkages, this project takes an interdisciplinary approach of combining genetic engineering with molecular modeling and chemical treatment, while leveraging machine learning and artificial intelligence. Using the model plant Arabidopsis, First-principles based simulations will be performed to model the formation of LCC linkages in the plant cell wall, to characterize their type and frequency and to evaluate their relative stability. Candidate genes controlling the linkages identified with in silico analysis will then be knocked out with genome editing or with RNA interference, with the aim to systematically alter the stability and propensity of the most stable linkages. The extent of genetic modifications will be controlled to preserve the basic structural stability of the plant and the yield of its edible component. The efficacy of this approach will be tested using molecular simulations, mimicking the solvent enabled chemical breakdown of biomass, followed by fractionation experiments. Going beyond this proof-of-concept, data mining will be implemented to identify optimal plant species, linkages and the associated genes for modification. Sustainability considerations will be included in the screening and identification of plants and linkages. The refractory nature of biomass has always impeded its fractionation. Since, this project takes a unique and preemptive approach of identifying and mitigating the linkages that create the resistance in the first place during the growth stage of the plant and then combines it with a solvent based treatment, we believe that it has the potential to transform the biomass utilization paradigm.

BACKGROUND. Bacteria form biofilms at physical interfaces in nearly every habitat of Earth's biosphere. Biofilms are structured communities of microbial cells enmeshed in a matrix of polymers. A fundamental problem in biofilm microbiology has been to understand how bacteria sense mechanical attributes of their environment and respond to build a biofilm. Progress has been made in understanding how mechanosensory apparatuses operate at solid-fluid interfaces; however, the sensory mechanisms operating at fluid-fluid interfaces - a key ecological niche for biofilm growth - remain unknown. One reason for this knowledge gap is the interdisciplinary complexity of methods required to study the problem.

OBJECTIVES/APPROACHES. Our overarching goal is to build disruptive methodologies to elucidate the principles of bacterial mechanosensing that drive biofilm assembly at fluid-fluid interfaces. To achieve this goal, the proposed project consolidates interdisciplinary expertise in rheology, colloidal physics, microbiology, biochemistry, and genetics. We will:

1. Identify the sensory apparatus(es) and signal transduction pathway(s) that drive the early stages of biofilm assembly at fluid-fluid interfaces.

We will use microscopy, optical tweezers and antibody-conjugated particles to manipulate and measure forces on single cells while monitoring pathway-specific bioreporters of biofilm-linked gene expression. We will quantify the sensitivity of sensory apparatuses and use bacterial strains with precisely engineered genotypes to disentangle potential roles for mechano-, rheo- and energy-sensing in cellular decision-making.

2. Chart the dynamic roles of mechanosensing and local micromechanics during biofilm maintenance at fluid-fluid interfaces.

We will develop imaging platforms, particle-tracking methods, and visualization algorithms to quantify and model the spatiotemporal development of micromechanics in biofilms while recording the emergent social behaviour of bacteria. We will use transcriptomics, gene knockouts, and CRISPRi to elucidate the sensory pathways that coordinate biofilm maturation and maintenance at fluid-fluid interfaces.

SIGNIFICANCE. Across all sectors, biofilms have a global economic impact in excess of $5 trillion each year.  Our outcomes will inform solutions for controlling biofilm development and facilitating its removal.  

Canadian arctic communities are at the tipping point of environmental change associated with climate warming. A major environmental change observed in this region is the thawing of permafrost (permanently frozen ground), leading to the degradation and subsequent collapse of the landscapes that support ecosystems and infrastructure. The implication of this major environmental disturbance extends beyond ecosystem and infrastructural degradation to include other critical and complex risk associated with medicine, food, and water, which are essential to human well-being. Indeed, permafrost thaw presents both human health risk and opportunities because the release of ancient biochemical materials into the environment could introduce novel disease-causing organism or antimicrobial compounds,  which could disrupt ecosystem processes, contaminate food, and water resources. Some microbes from degraded permafrost layers could exhibit antimicrobial resistance (AMR) characteristics when exposed to modern antibiotics. The proliferation of antibiotic-resistance is possible in degraded permafrost environment as novel microbes carried in meltwater encounters soil and buried antimicrobial waste from food, wildlife, and human health systems. This presents a dire risk to arctic ecosystem structure and human health, which constitutes a knowledge gap that needs to be addressed from the lenses of the traditional knowledge of indigenous people that live in this environment. A novel approach to address this gap involves  building on the environmental history by integrating traditional storytelling with recent advancements in environmental monitoring, chemical biology, bioinformatics, and metagenomics. The objectives of this research are to: 1) develop a conceptual model of arctic environment change through traditional storytelling; 2) assess potentials of novel antibiotics and AMR genes across landscape units that evolve from permafrost degradation; 3) identify ecosystem factors that interacts with the evolution of AMR and the feedback on ecological processes; and 4) evaluate the potential risk of AMR genes from novel permafrost environment to ecosystems functions, wildlife, and human health. The research approach will integrate traditional story telling with western field and laboratory-based research methods to generate empirical data, which will be applied to robust bioinformatic analysis and epidemiological modelling to inform public health measures in arctic communities. 

This project brings together experts in hardware and software design, bridging the digital-physical divide. We aim to apply machine learning (ML) models to develop state-of-the-art decision support tools for engineering computer-aided design (CAD). A key step in traditional engineering design, CAD modeling is increasingly important given the rise of digital simulation and manufacturing. ML techniques present the possibility to augment current modeling with data-driven recommendations and guidance.

Each 3D model designed in CAD - for example a car part or medical device - could be designed via a diverse set of user design sequences (e.g. sketch, extrude, constrain, fillet), some efficient and effective, and others not. For years, the CAD-research community has been unable to obtain adequately large user-datasets from which to draw generalizable conclusions about optimal design process; thus, much of the best practice around CAD design is heuristic, passed down by inefficient and inaccessible apprentice-style learning.

Recently, CAD moved to the cloud, thus unlocking broad access to user data; this project team has acquired a unique data set of >10 million professional cloud-CAD user actions. We take inspiration from the software design literature, where recurrent neural networks (RNNs) have been trained for recommender systems based on the large-scale open-source software code base (e.g., GitHub Copilot, GPT-3). We propose an interdisciplinary approach to analyze and improve the CAD design working efficiency through three objectives: (1) To qualitatively analyze the design workflow and augment the dataset of design sequences with explanation and rationale behind each step; (2) To train a predictive RNN that will recommend the next design activity; and (3) To evaluate the effectiveness and usefulness of the method in an educational setting. We will further develop new knowledge of human-AI interaction for computer-supported CAD design, such that our outputs are trustworthy and ultimately, implemented.

This project is high risk because it is among the first to combine ML models with CAD design user-data; despite our large dataset, the task is complex and thus there is uncertainty in our potential to train an accurate model. The predictive model has high reward potential, from accelerating CAD education and training, to reducing conflicts in CAD collaboration, to increasing efficiency of product development and manufacturing via lower rates of rework.

Globally, anxiety and postpartum depression (APD) are one of the leading causes of disability in women at the perinatal stage. In Canada, 23% of mothers have reported APD and among those only 32% have received treatments since the birth of their child in the year 2019 due to denial, stigma, poor mental health awareness, and the onus on the mothers to reach out for help. Currently, the first step and the only way of educating women about the risk, signs, and symptoms of APD is via questionnaire-based screening tools during their routine checkups. However, often personal barriers such as views of mental health in mothers significantly deter them from obtaining mental health diagnosis and treatment. This leads to poor quality of life and deteriorating emotional and physical well-being of their child and families.

To address these challenges, we will develop a new wearable mental health monitoring tool by taking an interdisciplinary approach combining biomedical engineering, human-computer interaction, material engineering, and maternal and child nursing. A biocompatible cellulose-based wearable device will be developed consisting of an array of sensors that can pick up electro-physiological, physical, and sweat biomolecular markers to investigate the diagnostic signature of maternal mental health. These signature-based diagnostics will utilize multiple markers in combination, effectively creating a `diagnostic fingerprint' of APD hence providing convincing and self-guided useful information to mothers about the predictors of their mental health. This project will also leverage the state-of-the-art computational optimization techniques to create a highly optimized, and compact wearable device packed with sensors that can pick up the wealth of multi-modal health data for creating such diagnostics fingerprints.

The new interdisciplinary approach that we will employ during this research will enable an understanding of the effects of perinatal mental health predictors on child wellbeing and will have the potential to dramatically improve mothers' health outcomes. Addressing perinatal mental health promotion, prevention, early intervention, and treatment of common perinatal mental disorders using these new wearable interventions will also be a step towards achieving Canada's Sustainable Development Goals 3 (good health and well-being), 5 (gender equality), and improving maternal, newborn, and child health outcomes.

Hirschsprung's disease (HD) is a life-threatening congenital disease where part of the patient's distal gut lacks enteric neurons needed to pass feces along. Current surgical resection of the aganglionic bowel can lead to various complications. Alternative research focused on cell transplantation of neural progenitors in the aganglionic colon but is still limited by cell expansion in vitro and long-term viability. We propose a "high-risk, high-reward" project, aiming to target the endogenous enteric neurons still present in the ganglionic bowel and stimulate their migration towards the aganglionic colon.

We recently discovered that post-migratory enteric progenitors from mouse Hirschsprung embryonic (E14.5) intestine can be induced to migrate by administration of glia-cell derived neurotrophic factor (GDNF), which is well-known to promote neural migration, or human amniotic fluid stem cells conditioned media (hAFSC-CM), which contains extracellular-matrix (ECM) re-organization proteins. Our innovative approach is to combine the stimulation of enteric progenitors with GDNF with the change in the ECM environment to increase permissibility for cell migration using hAFSC-CM to induce neural migration, as well as synergistically using the bioengineered materials to package and deliver the factors. We propose to test our hypothesis in the post-natal intestine and increase the translatability of this novel therapeutics from animal to human HD.

Objectives:

1. Utilizing an established bioreactor system for gut explant studies to investigate the effectiveness of GDNF and hAFSC-CM on post-natal Hirschsprung mouse colon in promoting endogenous enteric neural migration

2. To explore the method of delivery using the in vivo Hirschsprung mouse model, testing biomaterials combined with GDNF and hAFSC-CM that could induce a directed pattern of neural migration towards the aganglionic bowel

3. To investigate neural migration after administration of biomaterial containing GDNF and hAFSC-CM to human Hirschsprung specimen from surgical resection and maintained ex vivo using the bioreactor

By combining knowledge in developing biomaterials from biomedical engineers, together with the understanding of HD from pediatric surgeons, our "high risk" study objectives could result in the "high reward" discovery of a novel treatment for Hirschsprung's patients without invasive surgery.

Multiple sclerosis (MS) is a progressive and disabling disease of the central nervous system (CNS) that affects over 90,000 Canadians, especially women (75%), at the peak of life (20-40 yrs.), resulting in reduced employability and lifespan. Abnormal immunity, particularly CNS innate immune mechanisms, drive MS resulting in neurodegeneration. Epstein-Barr virus (EBV) infection is a putative cause of MS, although its relationship to MS onset and progression is uncertain. An emerging area that unifies immunity, viral infections and the CNS is the field of glycomics.

We propose to evaluate the hypothesis that EBV drives the MS disease process through induction of aberrant immune responses mediated by distinct glycomic profiles in CNS innate immune cells resulting in neurological disabilities.

Aim 1: Identify differential glycomic signatures and proteomic-associated profiles among persons with MS (PwMS) versus age- and sex-matched controls in the presence or absence of specific EBV immune responses within brain and serum.   

Aim 2: Develop new glycomic tools to define how viruses prime immune cells within the brain.

Aim 3: Establish in vivo models to explore a causative role for EBV in the development of MS.

Glycomic changes can regulate and prime the immune system, but little is known how this relates to neurological diseases like MS. We are affiliated with an MS clinic that serves over four thousand PwMS with an established history of integrated clinical and laboratory research partnerships. In Aim 1, we will define glycomic changes in human MS samples. In Aim 2, we will develop a tool set that allows us to examine glycosylation in conjunction with single cell transcriptomics in the brain. Application of machine learning to these datasets will yield detailed information on immune state in relation to the glycome. To investigate fundamental molecular and cellular underpinnings of how EBV can cause MS, we will establish novel mouse models in which mice are primed EBV infection in conjunction with subtle myelin injury (Aim 3). Our team, composed of a chemist, a viral-immunologist, a computer scientist, a neuroscientist, and a (MS) physician, brings expertise in immunology, virology, molecular arrays, machine learning, animal models, and clinical cohorts to bear on this important problem.

These studies will yield a deeper understanding of the molecular and cellular mechanisms that contribute to neurological diseases lighting the path to new therapies.

The heart is the first functional organ that develops in the human body. 1 in 100 children are born with congenital heart diseases (CHDs) that range from structural to functional defects and are the leading cause of birth-defect-related deaths worldwide. Currently only ~20% of CHDs can be linked to a known cause including genetic perturbations, teratogen exposure, and maternal viral infections. Specifically, the cellular and molecular dynamics of heart development that go awry in CHDs remain unexplored. There is therefore an unmet need to better understand the disease mechanisms of CHDs. Progress is hampered by the lack of a 3D model of human heart development that allows studying CHDs in a human context. We aim to establish a novel human pluripotent stem cell (hPSC)-based model that recapitulates the early events of heart development in 3D. To this end we recently adapted a gastruloid-based 3D model to the hPSC system that mimics gastrulation including the formation of all three germ layers and the formation of body axes. This model will enable studying the complex interaction of multiple cell types and the morphogenic events that occur during human heart organogenesis in an embryo-like context.

Here we are proposing to: 1) extend our gastruloid-based model to the early stages of heart development including formation of a cardiac crescent, heart tube, and primitive heart chambers; 2) apply state of the art genetic engineering, live imaging and computational analysis to track the migration of cardiac progenitors and their contribution to the different parts of the heart; 3) study hypoplastic left heart syndrome, a CHD resulting in the lack of a left ventricular chamber, using the novel 3D model to identify which cardiac progenitor populations are affected and at what point left ventricular development is interrupted.

Our team has extensive experience in modeling cardiac development with hPSCs and we have already generated axis patterned gastruloids that develop cardiac structures. We are therefore well positioned to perform this project. Our interdisciplinary team consist of experts in stem cell biology, cardiovascular development, live imaging, quantitative imagine analysis, biophysics, and mathematical modelling. Overall, our project will result in a unique 3D model of human heart development that will revolutionize our ability to study CHDs in a human context and provide novel insights into the disease mechanisms of CHDs. 

Imaging brain activity is fundamental in understanding cerebral function and how pathologies such as neurodegenerative diseases, stroke or epilepsy alters it. Functional Magnetic Resonance Imaging (fMRI) is the only broadly available approach that can image the whole brain and, while boasting multiple advantages, suffers from the requirement that the brain be imaged while the subject is in the unnatural MRI tube environment. Recently, a novel imaging method called functional ultrasound imaging (fUS) was shown to provide high sensitivity and can be used in awake, moving small animals. However, fUS is ill adapted to transcranial imaging, especially in larger animals and humans.

We have recently developed dynamic Ultrasound Localization Microscopy, which can map the super-resolved dynamic behavior of blood flow at the venules and arterioles scale by injecting, localizing, and tracking microbubbles (MB) approved for human use. Compared to fUS, dULM can image the hemodynamics at a smaller scale and, because of the enhanced signals from microbubbles, can be used transcranially even in humans. We have recently shown for instance that dULM can be used to quantify pulsatility through skin and skull in animal models.

The objective of this proposal is to develop, validate, and demonstrate initial feasibility of using dULM for the imaging of the brain activity.

Feasibility:

Aim 1) Develop a dULM-based brain activity biomarker and validate it against fUS

Aim 2) Demonstrate that dULM can be leveraged to quantify blood flow regulation in a mouse model.

Aim 3) Establish initial translational feasibility in a non-human primate model

High risk: If successful, this proposal would yield a unique functional dULM imaging method that could be used to reveal, for the first time at the smallest scale at depth, the complex and challenging spatial-temporal relationship between neuronal activity and blood flow in small animals and the brain function of large animals and potentially humans while they operate in natural environments. High reward: Functional dULM would significantly advance current knowledge of brain function by allowing its imaging in new settings and at a smaller scale, which could lead to new impactful discoveries in the management of brain pathologies such as neurodegeneration and stroke. 

Anthropogenic climate change causes widespread adverse impacts to human societies, and poses significant risks as the planet continues to warm. There is thus an urgent need to reduce CO2 emissions to Net Zero and develop alternative carbon-neutral technologies to power our society. If successful, our proposed high-risk project would contribute to both objectives. Inspired by photosynthesis, we will attempt to develop a new carbon-neutral method of producing "solar methanol" - an important raw material for use as a liquid fuel or as a feedstock for other chemicals - using a single-step process combining CO2 capture and reaction with water. 

Our concept is founded upon novel composite catalysts constructed with microporous zeolitic chalcogenides (ZChs), serving as active and economical materials for CO2 capture and activation. The regular intercrystalline pore systems of ZChs may accommodate a variety of photo and electrocatalytic and upconverting nanoparticles for further transformation of CO2 using solar energy. This would be the very first attempt we are aware of to construct open-framework ZChs tuned for optimum CO2 capture and photoelectrocatalytic transformation into carbon neutral solar fuels.

The photoelectrocatalysts for the CO2-to-solar methanol transformation will be constructed from molybdenum compounds (e.g. MoC, MoN, MoS2) conjugated with upconverting NaMF4 (M = Yb, Er, Tm) nanospheres, which convert NIR photons to higher energy visible and UV emissions. This trifecta of  i) a new microporous active support; ii) a photocatalyst, and iii) an upconverting species, is a completely novel approach, providing a unique strategy to support climate change mitigation and the technological development of alternative low carbon fuels. 

Our interdisciplinary partnership will leverage expertise in chemistry, and materials engineering, as informed by climate change mitigation policy, to develop a game-changing solar-driven process for the transformation of water and CO2, into a valuable energy carrier. Through this collaboration, our team will be able to confront the multi-faceted challenge posed by carbon-neutral fuels development. The impact of our proposed "CO2-to-solar methanol" technology on decarbonization and technological transition will be of a transformational nature, reaping significant societal rewards in terms of climate change mitigation and technological development of alternative fuels urgently needed today.

Breast cancer related lymphedema (BCRL) affects approximately 25% of breast cancer survivors. BCRL is a debilitating chronic condition characterized by a swelling in the arm that some survivors describe as being worse than the cancer itself. If diagnosed early, it can be efficiently managed. However, current diagnosis methods cannot detect the onset, are subjective and are not widely accessible.

With the increasing survival rate of breast cancer patients, development of accurate and accessible techniques for early diagnosis and objective staging of BCRL are urgently needed. To that end, we will exploit revolutions in point-of-care ultrasound (POCUS) and deep learning (DL) in three aims, as outlined below:

Aim 1: To determine biomechanical and other quantitative properties of tissue using POCUS in BCRL patients

Aim 2: To develop DL techniques to select high quality images and exploit raw ultrasound data for accurate and accessible diagnosis

Aim 3. To evaluate opportunities and challenges in access to POCUS and DL

POCUS is inexpensive and widely accessible. However, the exact composition of the tissue correlating with the various stages of lymphedema using ultrasound represents a major gap in knowledge.  We propose to quantitatively characterize these properties to facilitate diagnosis of early lymphedema.

Collection and interpretation of ultrasound images requires expert clinicians. The DL method will automatically select and analyze the highest quality ultrasound image, reducing the need for expert assessment. By combining advances in POCUS hardware and DL processing techniques, the proposed project enables accurate staging and early diagnosis with simple tests that can be performed even in remote parts of Canada.

Oncologists tend to refer patients with higher socio-economic status (SES) to lymphedema services at a higher rate. Moreover, geography continues to be a consistent influence on inequity in access to treatment services. POCUS and DL have the potential to improve the management of BCRL; but they can potentially exacerbate long-standing inequalities. In particular, privacy concerns as well as the black-box nature of DL affect trust. In this study, we aim to understand how SES as well as privacy and interpretability of DL affect the use of the proposed technology.

The proposed research will be carried out by an interdisciplinary team of engineers, clinicians, social scientists and patient investigators.

The future of work in Canada is changing. Many of these changes are in response to a rapidly evolving technological landscape. In anticipation of substantive workforce mobility and potential displacement in light of these changes, there is international momentum around (a) preparing graduates with appropriate competencies and (b) upskilling or reskilling the existing workforce. However, there is a third possible trajectory. Based on studies of workplace learning, this trajectory anticipates dynamics of innovation, transformation, and flexible decision making as workers incorporate new technologies into their day-to-day practices. This trajectory also anticipates that opportunities, risks, and rewards for learning and innovation are not equitably distributed. In essence, this third trajectory takes a critical look at learning at work and through work, even as work is changing.

To develop understanding of workplace learning and implications for the future of work, we locate our study within the domain of healthcare. Healthcare is a large, technologically rich industry that is poised for transformation and is of great public concern. In this study, we focus on health information systems (HIS) as a site to explore the cognitive, practical, and ethical dimensions of workplace learning when implementing new technologies. While HIS have been used for decades, new iterations of HIS are characterized by ever more digitization of the healthcare experience, are built to maximize the exchange of information within and across organizations, and include stakeholders from outside of healthcare to shape the design, implementation, and governance of these systems. Further, these HIS seek to harness the power of machine learning to help inform clinical decision making. These new iterations of HIS will change how people within healthcare think and work.

Our objectives are to build understanding about these transformations, their implications for work, and the required capacities for workplace learning. We propose a qualitative study based on interviews with healthcare workers and HIS designers combined with document analysis of key organizational and legislative texts related to HIS. This research is timely as it contributes to an evolving discourse about the future of work in Canada, takes a critical perspective to workplace learning, and considers the ethics of interacting with increasingly digitized workplaces in the public domain.

OBJECTIVES: This interdisciplinary project integrates psychology (SSHRC), health science (CIHR), and motor behaviour and computer sciences (NSERC) to explore how constructions of weight and body size stigma in virtual reality impact motor performance and learning. The research also examines body image factors as modulators.  Virtual realities offer a medium for embodiment in which people associate their virtual body with their real body. A virtual body has an impact on people's perception and behavior because they adapt to conform to what they expect their virtual body to perceive or do. In this way, manipulations of body size and weight in the virtual world may (1) identify implicit perceptions and behaviours tied to obesity stereotypes and weight stigma and (2) elicit motor learning and performance that translates to the real world.

RESEARCH APPROACH: In this experimental study, people will be randomly assigned to conditions where they embody smaller or larger body-sized avatars and perform motor learning and performance tasks. Based on stigma research, it is hypothesized that people embodying virtual larger bodies will show decrements in motor learning and performance. Based on our on-going interdisciplinary research, body image factors will strengthen the impact. Additionally, it is hypothesized that embodying avatars with different body sizes will provide people with the opportunity to act in non-conforming ways, which may actually work to counter stereotypes that are pervasive in society.

NOVELTY AND SIGNIFICANCE: As technologies continue to evolve, more time is spent working, learning, and playing in immersive virtual realities.  This work has important implications for fostering foundational motor learning and performance in this virtual world to transfer to the real world. Also, if embodying different body sizes can offset societal stereotypes, advocates for diverse and inclusive bodies can expand to the virtual world for heightened impact. The RISKS of this work are: it is unknown if realistic different sized bodies can be developed virtually, and if people will embody them; and the motor learning and performance tasks in virtual reality may not transfer to real-world. The REWARDS include being able to expand diversity of body types in virtual reality to counter stereotypes pertaining to obesity. Also, we can identify modifiable body image factors that can be targeted for intervention to strengthen the effects of virtual embodiment.

Per- and polyfluoroalkyl substances (PFAS) are a large and complex group of anthropogenic compounds which turn out to be ubiquitous persistent pollutants. Many PFAS are bioaccumulative and represent risks to humans and biota; they are particular concerns for northern indigenous communities that feed on arctic animals with high levels of PFAS. The most puzzling fact about PFAS pollution is that the identity of most organofluorine (e.g., >80%) in the environment and biota is unresolved. We propose to identify the PFAS `dark matter' based on new hypotheses: such PFAS are polymeric in nature, or they are incorporated as protein or lipid components. 

AIMS: We will integrate micro- and nano-plastics, proteomics and lipidomics research methods into PFAS measurement workflows to elucidate the identity of PFAS `dark matter' in environmental and biota samples. The goal is to resolve a long-standing scientific challenge and provide new knowledge to allow effective regulations to reduce the impact of PFAS.

RISK: All chemical analysis approaches, even the "gold-standard" liquid chromatography-mass spectrometry, have failed because the underlying hypothesis is that PFAS are low molecular weight and not incorporated into biological components. However, a large fraction of PFAS is used in polymeric forms, and many contain structures that resemble protein or fatty acids. Therefore, we propose a novel approach to analyze PFAS using the only instrument in Canada that combines pyrolysis-GC-MS with a high-resolution Orbitrap system. We will use the instrument's unique capabilities to find the PFAS that are otherwise not measurable and determine to what extent the pyrolysis interface allows the separation of PFAS occurring as polymers or possibly associated with lipids or proteins. This will provide a new approach to PFAS measurement. However, as these PFAS have unknown chemical forms, new approaches need to be invented, which are intrinsically high-risk, and high-reward efforts and often go unfunded.

TEAM: We have assembled a team uniquely positioned to identify the nature of PFAS dark matter: experts in analytical chemistry, environmental engineering, proteomics, toxicology, ecology, polymeric materials, and emerging pollutants. Our expertise is supported by a unique set of resources, including specialty mass spectrometry unique in Canada and access to contaminated environmental and biota samples, including the Canadian Arctic sensitive ecosystems.

The objectives of this research are threefold: to 1) develop novel community preference functions, which are critical components to 2) create a new optimization modeling framework for community decision- making and planning to 3) radically redesign the energy systems of Northern and Arctic communities facing extreme climate change. Doing so will integrate three traditionally siloed fields: behavioral economics, machine learning and energy systems optimization. This research is motivated by the need for advanced modeling efforts to support evidence-based decision-making, particularly models that endogenously incorporate the cultural and physical uncertainties related to climate change. Groups and individuals' specialization (e.g., farmers vs. fishers) can influence how they perceive and prefer energy generation. The heterogeneity within and across communities and complex climate change scenarios demand evaluation criteria representing the complexities of individual and community life. We elicit communities' economic preferences and beliefs with the use of field and choice experiments. Specifically, how people consider risk and uncertainty, intertemporal costs and benefits, and concerns for others in the community will inform the parameters of the community preference function. Unlike welfare functions based on efficiency maximization, the community preference function identifies social values and beliefs to estimate a welfare function capturing what communities value the most and aspire to (e.g., social cohesion, trust and cooperation). This function represents the trade-offs between alternative sustainable energy designs that communities support. The community preference functions are novel mathematical representations of community values. The functions will take the form of differentiable machine learning models which can be directly integrated into a stochastic optimization model. This project is high-risk due to the physical constraints of energy conversion - we may not be able to achieve the goal of a fully modular and adaptable design for a community energy system. However, it is high-reward because such a design would be an exemplar for communities globally facing climate change. Not only would such a system benefit the communities it serves, but it would also reap additional benefits for society writ large, in drastically increasing equity in energy access, addressing environmental justice and avoiding the large costs of abandoned energy assets.

Diabetic neuropathy is the most common complication developed in people with diabetes. It is characterized by the paradoxical combination of numbness and hypersensitivity that causes pain or burning sensations, usually starting in the toes and fingertips and advancing up the limb, due to progressive demyelination and regression of terminal sensory axons.

In a unique convergence of neurobiology, materials science, and regenerative medicine, we have recently discovered that a hydrogel, containing methacrylic acid (MAA) and known to induce blood vessel growth, also induces terminal axon growth under the skin, without growth factors or genetic modulation. We presume the underlying molecular mechanisms behind both the vascular and neural regenerative responses involve control of a non-fibrotic, alternative foreign body response, and preliminary studies of these unique neural effects point to the possible involvement of neuregulin1 (Nrg1).

Aim 1 is directed toward understanding how a material without any biological components can result in both neural and vascular regeneration responses in the host, and how the two responses are interrelated. Also of interest is the timing of axon growth in vivo (weeks) which does not coincide with the MAA-induced vessel formation (days). We will continue to use snRNA-seq, proteomics and novel imaging methods to identify molecular pathways of interest (e.g. Nrg1) and interrogate these pathways using appropriate inhibitors or knockout animals.

Aim 2 focuses on the ability to regenerate sensory neurons and normalize sensation in models of diabetic neuropathy. Initial studies will show that chronic hyperglycemia does not impede the neural regeneration induced by MAA (3D imaging, PCR, histology). Of prime importance will be demonstrating the functional consequences of MAA-associated terminal axon regeneration. Models of partial denervation and genetic disruption of neural support cells will verify that MAA can reestablish touch sensation without chronic pain (von Frey sensation test).

Together, aims 1 and 2 will guide us to improve the neural regenerative properties of the MAA-containing biomaterial and engineer it to reverse diabetic neuropathy in mice, and eventually those with diabetes. This treatment is unique in that it is growth factor and genetic modulation-free, achieved by a simple application of a hydrogel. It will provide a life-changing therapy for the millions worldwide living with diabetic neuropathy. 

Immunity is considered a non-transmissible individual trait except for the vertical transfer of immunity from mother to offspring. However, this vertical transmission is transient, and the immune system of neonates develops more enduring memory after the first encounters with pathogens and the microbiome. These pathogens and commensals are acquired from the environment but through interactions with cohabitants or other close contacts. Interestingly, these first encounters stimulate trained and specific immune memory. Therefore, if certain microbes generate consistent immune memory, the horizontal transfer of these microbes could lead to immunity that is transmissible between individuals within their social network. In this project, we will assess in a wildlife system how social networks influence the transmission of microbiomes and how this can escalate to the development of immune memory. For this, we will use South American fur seal (Arctocephalus australis) neonates as a model, since these animals develop stable social networks in their first weeks of life and their immune systems and microbes can be studied using tools from veterinary microbiology and immunology. We will repeatedly sample the upper respiratory system of these animals since their first day of life and monitor the changes of the respiratory microbiomes as their social network expands. To track their close contacts, we will develop a small tracking device that will generate data points when the device of other pup is in proximity. Using this data and through network modeling, we will determine if close contacts predict presence of specific pathogens and commensals and/or microbiome composition. Later, we will test the impact of specific microbe species or microbiome composition on the levels of secretory IgA and other immune proteins as well as the in-vitro activity of these proteins against respiratory bacteria (e.g. Streptococcus sp.) and viruses (influenza-A). Finally, we will model the impact of these transmitted immune profiles on the probability of infection with influenza-A and tuberculosis (Mycobacterium pinnipediae) when animals are naturally exposed to these pathogens during early adulthood (2-years-old). This study will provide a unique model for how social interactions can lead to immunity against respiratory infections. 

The major evolutionary transition from a single-celled lifestyle to multicellularity has occurred many times independently in the history of life, yet remains incompletely understood. The evolution of multicellularity has been examined experimentally, typically using a single species. However, individually species are increasingly considered inseparable from their microbiome. This project will use cyanobacteria and their attached heterotrophic bacteria (their microbiome) to look at the origins of multicellularity from a very different point of view. By exploring the nature of the interactions between large cyanobacterial cells and small heterotrophic cells, this project will raise a number of novel conceptual and theoretical issues: Can we think of multi-species colonies as a single unit of natural selection? Do all species benefit equally? If multispecies multicellularity is indeed observed, what does this mean for standard evolutionary accounts that focus on single species? These questions will be addressed by our interdisciplinary team including microbial evolutionary biologists, physicists, and philosophers of science. Colonies of the cyanobacterium M. aeruginosa and associated bacteria occur naturally in aquatic environments. Here we will recapitulate how colony formation evolves in the lab by allowing M. aeruginosa to replicate for hundreds of generations with or without different members of its microbiome. If successful, this two-year project will pave the way for an unprecedented long-term multispecies multicellularity evolution experiment. We hypothesize that multicellular colony formation will be favoured by associated bacteria that share a long evolutionary history in nature with M. aeruginosa, and that whole-genome sequencing over the course of the experiment will reveal co-evolution. Combining microscopy and microfluidics, we will then study the biophysics of how cells physically aggregate into colonies and describe the transition from individual to collective behaviours. Our results will provide insights into factors promoting and limiting single-cell interactions, revealing when and how collective behaviours could lead to obligate multicellularity. The theoretical implications for our understanding of what it means to be a multicellular organism will be profound, and on an applied level, understanding cyanobacterial colony formation may lead to novel strategies to combat toxic cyanobacterial blooms.

Depression is the leading cause of disability among women worldwide with 10-15% of perinatal women in the USA and Canada experiencing depression and anxiety during pregnancy or in the year following childbirth. Annual lifetime costs amount to over $45.9 USD and $20.6 CAD billion dollars in the USA and Canada, respectively. Psychological treatments are effective and preferred by perinatal women over pharmacological treatment. The COVID-19 pandemic has amplified the need for accessible mental healthcare and precipitated a dramatic increase in the use of telemedicine platforms. Some barriers can be addressed through task-sharing of treatment with non-specialist providers-individuals with no formal training in mental healthcare-and treatment delivery on telemedicine platforms. To date, an economic evaluation of these strategies compared to traditional specialist and in-person models has not been conducted. We propose to conduct an economic evaluation of an ongoing psychological treatment trial for perinatal depression which compares non-specialist providers (nurses and midwives) and telemedicine sessions to specialist providers (psychiatrists, psychologists and social workers) and in-person psychotherapy sessions in Toronto, Chapel Hill and Chicago (N=1368). We hypothesize that non-specialists and telemedicine will be less costly, equally effective, and associated with similar use and cost of other health services, and thus will be cost-effective compared with specialist providers and in-person sessions. Our results, which stem from the world's largest psychotherapy trial, have great potential to improve access to mental healthcare and to address the burden of perinatal depression and anxiety. If funded, this study will: 1) inform key decisions related to dissemination and scale up of evidence-based psychological interventions in Canada, the US, and possibly worldwide; 2) represent a chance for Canadian leadership in novel psychological service delivery models that will have global applications in both publicly-funded and for-profit healthcare systems; and 3) impact real-world practice by informing decision makers of the potential long-term savings to the larger healthcare setting in services to support perinatal women and their children. Our team includes world renowned interdisciplinary investigators alongside an established network of diverse stakeholders who will ensure the dissemination of findings at local, national, and international levels.

Primary Biliary Cholangitis (PBC) is the most severe type of liver autoimmunity, with an increasing incidence rate worldwide, including in Canada. Unfortunately, there is no curative drugs for PBC. Current treatment approaches are non-specific and poorly effective in 30 to 40% of the patients who continue to have disease progression.

Antigen-presenting cells (APC), a type of white blood immune cell, play a critical role in triggering a cascade of pathogenic events in PBC. They process and present small pieces (called peptides) of the self-proteins of liver cells (E.g., PDC-E2, a systemically-expressing mitochondrial self-antigen in PBC) on specialized MHCII molecules and, through which, activate another white blood cell type, CD4+ T cells. The activated CD4+ T cells then provoke other dangerous immune cells and mediate the destruction of liver cells, leading to scarring and permanent organ damage, necessitating life-saving liver transplant surgery. Thus, targeting APCs is the most promising disease-specific autoimmune treatment strategy.

High-risk and Interdisciplinarity

Using our vaccinology and genome engineering expertise, we will generate chimeric antigen receptors (CAR) against PDC-E2-peptide-MHCII molecules expressed on the APC membrane. We will engineer T cells to express this CAR (PDC-CART) to kill the APC and confirm the PBC amelioration by in vivo animal studies. To tackle the formidable challenges of generating CAR that must recognize both the peptide and MHCII, thereby getting practical applications for the disease treatment, the expertise from basic immunology, synthetic biology, and bioinformatics work in a synergy that would otherwise not be possible by any single researcher.

Transformative applications

1) Recent pioneering studies showed the generation of PDC-E2-specific CD4+ T cells by APCs not only in PBC but also in other hepatic and extrahepatic autoimmunity. Thus, CART cells specific for PDC-E2-MHCII molecule can potentially treat PBC as well as other autoimmune diseases, thereby revolutionizing the treatment of autoimmune diseases.

2) More broadly, a proof-of-concept that CART cells can be engineered for peptide-MHCII and hence in vivo immune modulation will enable corrective therapy across many highly intractable chronic diseases. For example, in autoimmunity, chronic infection, and cancer, one can modify CAR to MHC presenting peptides of self, foreign, or tumor antigens, respectively, to re-establish homeostasis. 

Crime scene interpretation has been criticized globally for subjectivity and lack of scientific rigor. Certain identification practices frequently rely on methods subject to human judgement, which are a leading cause of wrongful conviction. In one such area, bloodstain pattern analysis (BPA), existing pattern classification schemes have insufficiencies where they 1) misrepresent clarity concerning the cause and formation of the bloodstain; 2) are not conserved as the environment varies; and, 3) have issues dealing with feature ambiguity. Some branches of forensic science have modified existing schemes in response; but more recently, fundamental issues with the assumptions behind these definitions have been exposed.

Our research will take a fresh, data-driven approach, unladen by the assumptions in historical schemes. Our objective is to produce a next-generation classification scheme that integrates both expertise and a more accurate and informed understanding of the underpinnings of the causes of bloodstains. We will use unsupervised machine learning, a family of techniques that analyzes untagged data, and facilitates the emergence of natural patterns, which may otherwise remain hidden or obscured by human bias. This is completely different from supervised learning, already in use in BPA, where assumptions made regarding taxonomy form the basis of classification, rather than emerge from the data. Unsupervised learning has contributed to significant advances from medical diagnosis to marketing, revolutionizing what were manual processes based solely on expertise, experience and intuition. A wide range bloodstain patterns from indoor and outdoor environments will be used to train models (e.g. hierarchical clustering, autoencoders, Gaussian Mixture Models) to explore if pattern data can enable a similar revolution in data-driven forensic science. Since this is a new interdisciplinary approach, this represents a risk, yet has the potential to rewrite historical assumptions in determining salient features of bloodstains at multiple levels of granularity.

Our research will provide experts and stakeholders with a novel and scientifically rigorous decision-support tool to identify and explain the cause of bloodstains. It will enable broader use and accessible descriptions of crime-scene features to support and empower stakeholders on all sides of the criminal justice system.

BACKGROUND & OBJECTIVE: Ongoing pain, impaired function, or outright early failure are unacceptable outcomes following total knee replacement, one of Canada's most common and most expensive surgeries. Unfortunately, such issues are too common, with 1 in 5 patients expressing dissatisfaction after undergoing surgery. Quickly establishing good bone-implant integration and reducing joint inflammation during the early post-operative healing period is critical for long-term, pain-free artificial joint function. Recent studies in animal models and patient cohorts have indicated that bone growth and inflammation of the synovium (joint lining) are influenced by the gut microbiome, which are the microorganisms that live in the digestive tract. We propose a novel randomized controlled trial of a probiotic compared to placebo in patients undergoing total knee replacement. We hypothesize that patients receiving the probiotic will demonstrate reduced implant migration (indicating stronger bone-implant integration) and reduced synovial macrophage activity (indicating less joint inflammation) than patients receiving the placebo.

APPROACH: Probiotic or placebo therapy will be given from 1 week prior to through 6 weeks after surgery. Gut microbiome composition and encoded functions will be assessed pre-operatively in all subjects using metagenomic sequencing. Sample of bone and synovial tissue will be acquired intraoperatively to assess baseline composition through histology and immunohistochemistry. Surgery will be performed with a robot to optimize prosthesis implantation, with a prosthesis model that relies on bone ingrowth for fixation. Radiographic and functional imaging techniques will be used to measure implant migration between day 1 post-operation to 6 weeks post-operation, and to measure joint macrophage activity at 6 weeks post-operation. Patient-reported outcome questionnaires for pain and function will be acquired pre-operatively and at 6 weeks post-operation.

EXPECTED SIGNIFICANCE: If the probiotic is effective, this will be a paradigm-changing intervention in the field of orthopaedic surgery that currently only considers bacteria in terms of infection and has yet to consider microbiome therapies. The benefit of a probiotic therapy is that it that can be readily, broadly, and inexpensively translated into routine clinical practice across Canada and internationally. 

Background: Alcoholic beverages have been classified as a Group 1 carcinogen and are a leading risk factor for cancer (Rumgay et al., 2021). However, public awareness about the health risks associated with alcohol consumption remains low (Scheideler & Klein, 2018). Alcohol warning labels are recommended by the World Health organization as a mechanism for increasing consumer knowledge about the negative consequences of alcohol (WHO, 2018). Despite support for alcohol warning labels among Canadian's (Vallance, 2018), their implementation has been strongly, and successfully, opposed by the industry (Stockwell et al., 2020). The current study proposes working in partnership with the alcohol industry in hopes of developing warning labels that simultaneously increase consumer knowledge of alcohol-related risks and minimizing the negative impact on business practices. 

Aims: Our team is comprised of scientists from the fields of psychology, public health, nutritional science, and marketing, and an industry partner. Together this interdisciplinary team will (1) co-design and produce a set of alcohol warning labels, (2) pilot test the labels with consumers, and (3) test the effectiveness of these labels on consumers' attitudes, risk perception, acceptability, and intention in a real-world setting.

Method: Label designs will be qualitatively piloted using consumer focus groups. Focus groups will assess consumers perceptions and acceptability of the labels and refine the label design. A quasi-experimental design will then be used to test the effectiveness of the labels in a real-world setting. Pre-post surveys will be conducted with product consumers 19+ years from two industry partners (intervention and control site). Participants will be recruited at point of sale and via social media.

Significance: Few real-world studies have tested the effectiveness of warning labels. Labels have largely been developed by researchers without stakeholder consultation and have been tested online or in laboratory settings (Kokole et al., 2021). This interdisciplinary partnership seeks to overcome some of the barriers that are impeding the implementation of alcohol warning labels in Canada. Our innovative approach will determine the viability of working with an industry partner to enhance health knowledge among consumers, develop a set of novel warning labels, and build on the sparse evidence supporting the use of warning labels as an effective public health measure.

T-cell acute lymphoblastic leukemia (T-ALL), a malignant neoplasm of immature T cells, is a particularly aggressive leukemia with diverse subtypes lacking available targeted therapies or immune interventions. Large-scale sequencing efforts examining the genetic underpinnings of ALL at diagnosis and relapse have failed to inform effective targeted treatment strategies or predict relapse. As tumors evolve under the pressure of immune surveillance and immune dysregulation often precedes malignancy, immunotherapy has been proposed as an alternative to these anti-cancer drugs. Unfortunately, despite its success in a number of tumors including B-cell ALL, immunotherapy has not improved T-ALL patient outcomes. This is partly due to a lack of understanding of which immune populations interact with leukemia. It is also unknown how leukemic heterogeneity influences a patient's immune recognition. To elucidate this heterogeneity, we conducted a preliminary multiomic analysis, combining highly multiplexed protein marker detection with unbiased transcriptome profiling and T cell receptor (TCR) sequencing, of 36,714 cells from primary T-ALL (n=12) adult patients at disease diagnosis and healthy bone marow donors (n=4). This analysis revealed 1) high intratumoral leukemic heterogeneity, and 2) clonal expansion of cytotoxic T cells in leukemic patients. Based on these preliminary data, we hypothesize that the bone marrow immune microenvironment plays a role in supporting leukemia survival, progression, and escape from treatment. To address this hypothesis, we propose the following Aims:

Aim 1: Map intratumoral leukemic heterogeneity across data modalities

Aim 2: Identify TCRs specific for leukemia antigens

Our proposed studies represent the first comprehensive mapping of the immune system in primary human T- cell acute leukemia. This work will uncover novel subpopulations and cellular interactions that can be targeted to enhance treatment response and prevent relapse. Our team is uniquely positioned to perform these studies. We have extensive experience in single-cell transcriptomic profiling of the bone marrow, have previously described the T-ALL bone marrow microenvironment in animal models, and have categorized T-ALL heterogeneity in cell cultures at single-cell resolution. Our studies - combining in vivo models of T-ALL and single-cell analyses of primary patients' disease - have transformative potential to reveal new immunomodulatory strategies for T-ALL treatment.

Objectives- We propose to characterize the radiofrequency (RF) signature of artificial tissues grown in paper chips over a broad frequency range (50 GHz to 1 THz). A model will be designed to explain the data and link the RF signals to dielectric properties and biology.

Motivation- Biological phenomena are complex and hard to elucidate. Bioanalytical strategies can show limitations, such as cost, use of potentially harmful reactants, or high processing times. Innovative modalities are also likely to bring new perspectives. In this context, low power RF techniques offer non-ionizing, fast and low-cost alternatives to the state-of-the-art. However, the electromagnetic signature of biological materials is largely unexplored, especially in the THz domain. The wave-matter interaction is complicated by the difficulty to isolate homogenous tissue samples.

Methodology- A custom free-space setup will be built and modified to accommodate tissue culture (i.e. temperature, CO2, etc). Cell-based artificial models will be built by growing cells, in biogels, on chips made of filter paper. Vascular growth and muscle formation will be emulated on these chips. These phenomena involve large modifications of cell walls, which are suitable to electromagnetic detection. The RF signature of these reproducible, controlled biosamples will be recorded. A physical model, linking the signal to the known biology of the grown cells will be established.

Novelty and significance- This research aims at building a new way to study living matter. It will complement the state-of-the-art to provide a more comprehensive picture of life. This project offers unique opportunities to design and use a new technique that gives access to physical and biological parameters not routinely considered in biomedicine.

High risk- The multidisciplinary project will bring together two expert teams to combine cutting-edge RF and biochip technologies. The specificities of each aspect must be considered to ensure the relevance of the data. The ultimate task of the project (i.e. carefully explaining and linking the RF signals to the physics of bioprocesses) is a largely uncharted territory.

High reward- A non-ionizing bioanalytical system will be built. This first of its kind approach will complement standard tissue study, mostly based on routine biochemical methods, by describing previously unconsidered biological parameters. The project will put forward a new way to quantify living matter.

Objective: To implement a new outreach program that will identify, contact, assess and refer patients at high risk of kidney failure who are currently not receiving appropriate kidney care.

Background: Approximately 4 million Canadians have chronic kidney disease (CKD). It is the 10th leading cause of death among Canadians, with extensive impacts on quality of life for those living with advanced disease. CKD is largely asymptomatic until kidney failure is imminent, making it a hidden health problem that is frequently unrecognized by patients and physicians.

The Problem: Despite efforts to improve the recognition and screening of CKD, late referral to nephrologists remains common. In Canada, over one-quarter of patients with kidney failure see a nephrologist for the first time within 90 days of starting dialysis. Many patients are not receiving evidence-based kidney-protective therapies that could change the trajectory of their disease had they received nephrology care in a timely fashion.

Innovation: KidneyCare Outreach is a new program that will act as a safety net for Ontario residents with advancing CKD who are not receiving specialized kidney care. We will use routinely collected data from our publicly-funded healthcare system to provide targeted outreach for patients at high risk of kidney failure. This data is housed at ICES and our project represents the first time ICES data sources will be leveraged for direct patient contact. We have obtained Research Ethics Board approval and our approach complies with Ontario's Personal Health Information Protection Act.

Collaboration: Our interdisciplinary team includes nephrologists, lawyers, experts in health law and privacy, epidemiologists, and patient partners.

Impact: Our program will address a major care gap for patients with deteriorating kidney function who are not receiving appropriate kidney care. Our pilot study will generate critical preliminary data on the Program's impact on patient care. This will inform the design of a nationwide randomized clinical trial to definitively ascertain if our Program delays the onset of end-stage kidney disease and kidney disease-associated morbidity and mortality. Finally, it will guide pioneering legislation that will enable the judicious use of personal health information to enhance the delivery of care.

The recent unprecedented proliferation of digital technologies has drastically transformed human behaviors, business models, and social structures. With ongoing evolution of 5G, AI and Internet of Things (IoT), we are quickly leaping forward to a hyper-connected society, where ethical and trusted data exchange among humans and machines becomes the critical foundation.

The objective of the NFRF project is to innovate ethical and trusted data governance technologies that maximize human benefits in the era of human-machine coexistence. Specifically, our proposed research and innovation activities are focused on the following Themes:

a) Translating ethical and legal requirements into human-centred digital innovations. This Theme is dedicated to the identification and translation of human-centric ethical and legal norms/principles into design requirements of IoT data handling procedures. These requirements will be further embedded into new digital innovations aligned with policies, ethical and legal restrictions for trusted data governance.

b) Creation of new verifiable data handling mechanisms for trusted data governance. New accountable data governance procedures with defined security, trust, and privacy provisioning in connected systems that guarantee the preservation of key ethical and legal requirements will be designed. Ground-breaking technologies on data handling process verification, auditing,  ethical/legal rights & ownership protection, and privacy & trust management will be created for sensitive data collection, processing and sharing.

c) Knowledge mobilization and technology transfer for human-centric value maximization. To maximize the transformative impact of the NFRF project, new business models, processes and vertical applications will be created. By leveraging the innovation results of Theme a) & b), new medical, consumer and industrial applications that will bring a fundamental impact on our lifestyle, mental health, behavior and business will be developed.

By focusing on EDI-oriented training, this NFRF project will unify all critical multi-disciplinary expertise for ground-breaking research innovation on trusted data governance. The proposed research will create "from principle-to-practice" capability, which will bring tremendous innovation opportunities. Expected results and business development activities will empower both consumers and governments to safeguard their digital rights and policies in the hyper-connected world.

BACKGROUND

Lung cancer is the leading cause of cancer death. Immunotherapy (IO) can significantly increase survival for some patients, but with high cost and potentially severe toxicity. PD-L1 immunohistochemistry is used to predict for IO benefit but is insufficiently accurate. Tumour mutational burden (TMB) is a promising adjunct biomarker to determine which patients will benefit from IO, but costs >$1,000 CAD and requires one month to obtain, challenging clinical implementation. It has been shown that genetic data can be estimated from standard-of-care hematoxylin and eosin (H&E)-stained pathology slides. We will develop a system addressing the unmet need to quickly and inexpensively predict TMB from scanned H&E slides, to support the clinical decision for IO.

OBJECTIVES

1. To develop a deep machine learning system to determine the most TMB-relevant regions on scanned H&E-stained slides, and predict the patient's TMB.

2. To form a 3-way link between the pathologist's eye-gaze tracking during review of the system's relevance map on the H&E, the pathologist's assessment of qualitative patterns at each gaze location synchronously acquired via natural language processing, and the system's spatial prior, enabling explainability and iterative training.

3. To conduct a user study to determine the optimal interface configuration to aid the pathologist in interacting with the deep learning system and interpreting its results.

4. To conduct a prospective pilot study testing the feasibility of implementation of our system into the clinical workflow.

APPROACH

We will develop this system using cross validation on the squamous cell cancer (SCC) histology images (N = 1612 slides) and TMB values in The Cancer Genome Atlas public repository. We will test on 50 local patients with lung SCC and known TMB. We will also test prediction of IO outcome on a local 150-patient SCC data set involving lung and head and neck cancer.

NOVELTY AND SIGNIFICANCE

Our hybrid human-machine approach to machine learning explainability and iterative learning is novel in the setting of digital pathology, as is our application to TMB estimation for squamous cell carcinoma. Since H&E slides are routinely obtained, this approach will be fast and inexpensive. This will therefore enable accurate selection of an optimal systemic chemotherapy and/or IO regimen, improving treatment and minimizing the risk of severe side effects for patients least likely to benefit from IO.

The effective utilization of space resources is a key component to the long-term exploration of the Moon and other celestial bodies. International efforts are underway to setup a permanent lunar base as a stepping-stone for future space exploration missions, which requires the development of technologies for in-situ space resources utilization (ISRU). This exploratory research program will develop and evaluate a process to transform lunar regolith (moon dust) into metal and subsequently a solid fuel for powering in-space energy production and propulsion.

The lunar regolith abounds with metallic compounds, silicate, and metal oxides. Although chemically inert, we want to leverage our long-standing expertise in combustion, nanomaterial and sustainable energy to develop a process to produce metastable intermolecular composites which will be formed from the regolith derived metal particles and a solid oxidizer. This fuel undergoes a strong, yet highly tuneable, exothermic reaction that can occur in absence of external oxygen and is suitable for space use. 

This research project first seeks to develop a feasible thermochemical process for extraction of metal and allay from the lunar regolith with the aid of concentrated solar heat. Micro-sized energetic metal powders will be then fabricated via ball-milling and mixed with a variety of nano-sized solid oxidizers. The micro-structure, thermochemical stability, energy release and combustion properties will be characterized. Process conditions and experimental data will be collected from measurements in a vacuum thermal chamber and utilized to assess the energy and environmental impacts.

This work brings together distinct interdisciplinary expertise in planetary geology, energetic materials, aerospace propulsion and life cycle assessment of space resources. The effective transformation of the abundant lunar regolith into a compact source of thermal and chemical energy will have a disruptive impact on the lunar program and can serve as the basis for the in-situ utilization of space resources as a fuel for future interplanetary travel.

Rare metabolic diseases represent a unique challenge for human health. These diseases can manifest with serious clinical symptoms such as heart disease, but it is typically very difficult to link patient symptoms to an underlying biological mechanism, making discovery of new treatments problematic. Our ability to explore the cause and progression of rare diseases is continually hampered by a lack of adequate experimental models. Heart tissue samples from rare disease patients are incredibly scarce, and experimental models created from other cell types might not fully recapitulate the functional defects in tissue metabolism and muscle contraction. Animal models cannot be generated without a fulsome understanding of the disease. Finally, it is highly unethical to test treatment strategies directly on the patients without extensive laboratory testing.

To address this knowledge gap, our multidisciplinary team will combine cutting-edge three-dimensional (3D) bioprinting technology with human stem-cell culture to create patient-specific experimental models of rare metabolic diseases. Using a `bedside to bench' approach, we will leverage well-characterized rare metabolic disease patients with identifiable clinical phenotypes and genotypes to yield patient derived induced pluripotent stem cells. These cells will be differentiated into heart muscle and incorporated into 3D bioprinted structures using a unique `one click' process, that will allow our methods to be adopted by laboratories with minimal tissue engineering expertise. The resulting heart tissue is easily handled, can be exposed to external stimuli using off-the-shelf cell culture and organ bath equipment, and can readily be assessed for contractile and metabolic function.

Where most tissue engineered heart models aim to create fully developed adult tissue, we will distinguish ourselves by understanding the conditions required to replicate the dysfunction seen in human rare disease patients. We will then use our model to test candidate therapeutics that recover mitochondrial and metabolic function, to establish their suitability for improving heart function in human patients. Collectively, this project will reward us with increased research capacity for studying rare diseases, and an improved understanding of the pathogenesis and treatment of metabolic diseases that affect the heart, which will improve the lives of people in Canada.

In this proposal, we will develop methodology to identify proteins carrying the post-translational modification oligosialic acid, a glycan thought to be important for cancer progression.

Half the population of Canada will be diagnosed with cancer in their lifetime and another half of those will die from the disease, making cancer the leading cause of death in Canada. A universal hallmark of cancer is altered cell surface glycosylation, with altered sialylation being one of the most distinct changes. Progressively higher levels of sialic acids strongly correlate with poor prognosis. Elucidating the mechanisms by which hypersialylation drives cancer progression is a growing area of research, particularly with regards to sialic acid-mediated immune suppression.

Sialic acids come in several varieties - monoSia is one sialic acid sugar capping the end of an underlying glycan, oligoSia is a short chain of 2 to 7 sugars, and polySia is comprised of a linear chain of 8 to more than 400 sugars in length. MonoSia and polySia are both well studied, in large part due to the abundance of analytical tools with which to study them. However, a lack of good analytical tools has left oligoSia virtually ignored - the proteins which are oligosialylated have not been identified, the biosynthesis of the glycan has not been resolved, and the role of the glycan remains largely unknown.

We propose to develop methodology that will allow us to identify oligosialylated proteins. Our methodology will involve enzymatically labeling oligoSia with a chemical tag that can then be used to identify proteins using mass spectrometry. We will use this methodology to identify oligosialylated proteins in serum and cell lysates from healthy humans as well as those with cancer. Our research will provide a platform for our investigation of oligoSia in cancer. Additionally, our methodology can also be used by the research community to study neurodegeneration, mental illness, and autoimmunity, processes in which sialic acids are also known to play pivotal roles.

Background: In North America, Lyme disease (LD) is caused by the bacterium Borrelia burgdorferi and is transmitted to humans and other mammals by infected black-legged ticks. Symptoms include a characteristic skin rash called erythema migrans. If untreated LD can lead to serious medical outcomes including arthritis, carditis, and inflammation of the brain and spinal cord. In the USA approximately 476,000 individuals are diagnosed and treated for LD each year. While the numbers are far fewer in Canada, Borrelia burgdorferi infected ticks, and reported LD cases, are increasing annually.

Individuals presenting with erythema migrans, and have been exposed to black-legged ticks, are treated with an extensive course of antibiotics. In some cases, additional highly sensitive and specific serological diagnostic tests can be obtained 4-8 weeks post infection.

Unlike serological tests, a direct LD test would: a) detect active LD infections regardless of previous exposure; b) inform how to treat an individual who has been bitten by a tick but does not present with other typical symptoms; and c) serve to monitor the success of treatment. However, despite much effort, even the best results from such tests have relatively low sensitivity (<60%) and are thus currently excluded from LD diagnostics.

Objective: Develop a completely new direct test for Lyme Disease.

Approach: Our novel direct-testing method connects recent technological advances in liquid biopsies used in cancer research to the prokaryotic restriction modification system that protects bacterial DNA by depositing the epigenomic modification N6-methyladenine.  Importantly, N6-methyladenine has recently been proven not to exist in mammals. Our approach begins by enriching for N6-methyladenine by immunoprecipitation (IP) prior to down-stream detection. We will first optimize the N6-methyladenine-IP assay using cell and animal models followed by highthroughput DNA sequencing and computational analyses. We will then move to the testing of clinical material.

Expected significance:  A direct diagnostic test for LD would allow for an early diagnosis to patients, including those with previous LD infections.  By design, our proposed early diagnostic test would work for different LD-causing bacteria and would complement existing clinical workflows. Our universal clinical-grade direct LD test would provide patients with an unprecedented means to detect LD and receive the appropriate treatment.

Hurricanes, threatening Atlantic Canada almost annually, directly affect millions of residents while indirectly exposing millions more to extensive hurricane media coverage, triggering cognitive, emotional, and behavioural responses with both short-term and longer-lasting mental health and wellbeing impacts. Current studies primarily utilize a post-hurricane retrospective approach, threatening "perishable" or "ephemeral" data collection (before memories fade and physical evidence is erased) thus jeopardizing a nuanced understanding of dwellers' intense ongoing adjustments occurring across a complete hurricane cycle, namely pre-, peri-, and post-hurricane.

Responsively, this project employs a novel rapid response approach to examine Atlantic Canadians' cognitive, emotional, and behavioural responses to hurricane media coverage and the coverage's mental health and well-being impacts across a full hurricane cycle through three tasks: 1) a large-scale, rapid-response, three-wave quantitative survey: pre-hurricane (two weeks before hurricane landfall), peri-hurricane (96 hours after hurricane onset), and post-hurricane (two weeks after hurricane landfall); 2) post-hurricane recovery in-depth qualitative interviews; and 3) key stakeholder action workshops to translate research outcomes to support community-based practices. Merging natural hazards (hurricanes) and media and communications with psychology and cognitive sciences, this project is innovative and significant as the first Canadian effort to conduct a full-hurricane-cycle assessment of dwellers' hurricane risk perceptions, health protective behaviours, and acute adjustment stress.

Hurricanes are inherently unpredictable, making designing a rapid response study to capture coping responses, spanning from warning siren to long-term recovery, very challenging. Traditional research processes (i.e., the grant administrative cycle and research ethics review) delay rapid research team deployment and participant recruitment. This high-risk project, breaking disciplinary, methodological, and administrative boundaries, reaps high reward as time-sensitive data will 1) support community-based mental health and social care organizations to develop customer-centric services; 2) encourage public media to adjust their information delivery models to avoid potential negative public health impacts; and 3) inform disaster and emergency management, and other sectors, to improve public disaster education.

Ammonia, a multi-functional chemical used in the food, energy, and trade sectors, has a unique role in a hydrogen (H2) economy that uses H2 and its carriers as energy vectors or "currency" for energy production, storage, distribution, and utilization. About half of the H2 produced globally is converted to ammonia, the majority of which is used as base materials for fertilizers to increase food production. Ammonia is also an efficient H2 carrier that has more energy per cubic meter than pure liquid H2 under engineering conditions. It can be directly used in engines or fuel cells to power the energy sector, which might create an unhealthy competition against fertilizer uses for food production in a scaled-up H2 economy. In addition, ammonia is distributed internationally in stainless steel tanks on land or at sea at a low cost. Therefore, trading ammonia as a H2 carrier can secure energy export channels for Canada to a growing global H2 market.

The goal of this project is to enhance our understanding of low-carbon ammonia's role as a "currency" to connect the food, energy, and trade sectors dynamically without negative competitions, and develop a novel scalable electrochemical power-to-ammonia (P2A) process that emits zero carbon to fit into the multi-sectoral dynamics. This project will accelerate the transition to a H2 economy that can help achieve the net-zero emissions target in Canada, while creating new jobs and promoting economic growth as identified in the Hydrogen Strategy for Canada. Our interdisciplinary team will combine expertise in economics, electrochemical catalysis, heat and mass transfer to meet the following objectives:

1) design a scalable mild-condition solid-state P2A reactor (significantly lower temperatures and pressures than conventional) with zero carbon emissions,

2) study the P2A reactor with intensified electric heating and beneficial waste heat recuperation to lower the electricity consumption at different output scales, and

3) quantify the economic benefits, risks and constraints associated with using low-carbon ammonia in multiple sectors impacted by factors such as taxation, sector competition and environmental policies.

This is a truly interdisciplinary approach: the technology development will provide inputs, e.g., electricity consumption and efficiency to the economic analysis, while the economic analysis will guide the technology development with targets such as optimal production scales in different sectors. 

Manufacturing metaverse is an emerging concept of digital transformation of production systems by creating persistent and digital representations that are connected to the physical entities like people, product, and process. It allows physical-virtual interactions and testing environments for finding more sustainable and resilient solutions. To build a manufacturing metaverse, the key is to create digital twins (DT) of all components of the physical system with full knowledge of its behavior and characteristics. Current DT technology has provided viable solutions for modeling product and process while rare progress has been reported for human DT (HDT). Existing virtual human representations like avatars use average anthropological data (e.g., height and arm length) to support ergonomic studies. However, such low-fidelity HDTs cannot meet the need of providing accurate simulation for customized human operations in dynamic settings and supporting collaborative intelligence such as proactive human-robot collaborations where robots understand human intent and respond empathically and adaptively.

Our collaborative team includes expertise in manufacturing, computational science, robotics, cognitive science, and psychology. We aim to develop a characteristics-rich HDT for supporting high-fidelity simulation for human-in-the-loop manufacturing systems. The enriched HDT will be featured by personalized human attributes (e.g., emotion states). The overarching goal can be achieved by 1) building a unified HDT model with an open architecture that supports easy adaption and expansion, 2) developing deep learning algorithms for predicting operator cognitive behaviors based on multi-channel information (e.g., gait and facial images), 3) associating operator behavior with safety and health risks (e.g., fatigue) in manufacturing settings, and 4) establishing novel robot training strategies for proactive human-robot collaboration based on HDT simulation.

The proposed research would considerably push the boundaries of human digital twin modeling, simulation, and human-machine collaboration toward a futuristic manufacturing paradigm with minimized trial-and-error planning efforts and maximized operator wellbeing awareness. The developed characteristics-rich HDT can be adapted to other fields beyond manufacturing such as healthcare, social behavior study, and wearable product customization with broad social and economic benefits.

Heart failure (HF) is the No. 1 cause of death globally. Around 26 million people suffer from HF each year and about 50% HF patients die within 5 years.  HF occurs when conditions such as myocardial infarctions (MIs) or hypertension-related cardiac hypertrophy result in decreased oxygen flow to the heart tissue, causing cardiac muscle cells (cardiomyocytes) to die. Promoting cardiomyocyte survival and/or regeneration is one of the most important strategies of HF therapies. However, most treatment strategies target symptoms. There are currently no clinically approved agents that promote cardiomyocyte survival and/or regeneration.

LATS is a protein kinase and tumor suppressor that inhibits cell proliferation and induces cell death. It was recently discovered that LATS is upregulated during HF, and genetic inactivation of LATS inhibits cardiomyocyte death induced by MI in animal models. We hypothesize that selectively inactivating LATS in the heart using small molecule (SM) inhibitors represents a novel strategy for HF therapy.

The traditional process of developing a targeted therapeutic agent is labor intensive and very time consuming. In this study, we propose a new method that combines artificial intelligence (AI), novel biosensor and organ-specific drug delivery technologies for cardiac-specific targeting of LATS kinase for HF therapy.

To achieve our objective, we will first employ artificial intelligent (AI) to identify SM candidates that show specific binding to LATS. We will then test the candidates for effective inhibition of LATS in vitro using a LATS biosensor. Next, we will evaluate the ability of the LATS SM inhibitors to promote cardiomyocyte survival and protect cultured cardiomyocytes from hypoxia (ischemia-mimicking)-induced cell death. Third, we will deliver top LATS SM inhibitors specifically to the heart using a new cardiac-targeting SM Nanobubble technology and evaluate their protective effects on MI-induced HF in a mouse model.

Our team is comprised of experts who have documented success using AI in the drug discovery process, have developed cancer drugs using state-of-the-art bioluminescent biosensor technology, have expertise in SM drug synthesis and delivery, and have extensive backgrounds in basic and clinical cardiovascular research and pathology. This uniquely interdisciplinary team is best positioned to apply cutting-edge cell biology knowledge learned in cancer research specifically to the heart for HF therapy.

Disease: Over the course of a lifetime, 1 in every 8 women will develop invasive breast cancer. Skeletal integrity becomes increasingly compromised in women after menopause and further affected after breast cancer diagnosis with related treatments. Osteolytic bone lesions that are prevalent in metastatic breast cancer patients weaken the skeleton and result in painful skeletal-related events that are difficult to treat or can even be fatal.

Challenge: 1) Exercise, a non-drug intervention with proven health benefits, is commonly prescribed. Exercise, however, is challenging for elderly patients because of multiple factors including osteoporosis/sacropenia with aging, fear from activity related fracture, and/or fatigue from cancer treatments. 2) Circular RNAs (circRNAs) have been shown to regulate breast cancer progression. However, it is unclear whether circRNAs play essential roles in exercise mediated bone metastasis.

Strategy:  Whole body vibration stimulates the musculoskeletal system. Clinical trials on osteoporosis and cancer patients have confirmed the safety of vibration therapies in humans. However, effects are attenuated in postmenopausal subjects. To boost the skeletal benefits of vibration in elderly breast cancer patients, we will test if Yoda1-a highly selective activator of Piezo1 mechanosensitive ion channels-could act as an enhancer for the vibration effects.

Hypothesis: 1) Yoda1-augmented vibration by acting on Piezo1 can serve as an exercise alternative to suppress cancer bone metastasis on-chip. 2) Vibration activated Pizeo1 is mediated by its binding to circRNAs that bind other cytoskeletal proteins.

Objectives: 1) To investigate the effects of Yoda1-augmented vibration on cancer cells and bone cells using our novel microfluidic bone-on-chip platform that recapitulates cell-cell cross-talk in the bone microenvironment; 2) To identify the circRNAs that bind and mediate Piezo1's activation.

Impact: If circRNAs are confirmed to be critical in regulating cancer bone metastasis under vibration, new therapeutic approaches can be developed for metastatic breast cancer patients.

High risk: This will be the first study to examine roles of circRNAs in the mechanical regulation of osteocytes and cancer cell bone metastasis.

Interdisciplinary: The proposed project requires expertise on 1) Engineer: Vibration, microfluidics chip design, Organ-on-Chip; 2) Biology: circRNA biology; and 3) Medicine: orthopedic bone metastasis.

The conventional methods of particulate plastic pollution identification are primarily based on the interactions of photons with samples collected from water bodies. The detection and classification of the particles are usually performed in a lab with expensive equipment such as Raman spectroscopy, electron microscopy and gas chromatography. The process of collecting, transferring, and processing the samples is a major bottleneck in comprehensive understanding of particulate plastic pollution. We propose a novel approach that not only accelerates the process of detection and classification of microplastics but also is a magnitude order cost-efficient compared to current techniques.

Ultrasonic imaging is commonly used in noninvasive medical and industrial testing to detect invisible flaws in internal body tissues or industrial products. It is based on emitting ultrasound pulses and processing the intensity and the return time of the sound echoes. The intensity of the returned signal is a function of the type of material that reflects the sound. It is well-known that the composition and density of microplastics is different from the surrounding biota. Therefore, we propose a novel idea that ultrasonic imaging is an efficient method for detection and classification of microplastics. The significance of using ultrasound images is that 1) the analysis via ultrasonic imaging can be done in-site and there is no need to transfer the samples to the labs, and 2) it is significantly less expensive than laser and light scattering analysis methods. The major risk associated with this technique is the expected sensitivity and accuracy of the results. To deal with these risks, we plan to conduct experiments with various ultrasound sensors/probes, and enhance the robustness of the results by exploiting artificial intelligence techniques in the detection and classification processes. Particularly, we will apply instance segmentation algorithms such as Mask-RCNN to ultrasound images, and we will refine such detection algorithms by providing ground truth results from Raman spectroscopy and electron microscopy methods. The proposed technique will be integrated with underwater robotics systems to accelerate research on microplastics.

This proposal assembles a team of experts across ordinarily distant disciplines of analytical spectrometry, clinical pathology, data sciences as well as neurosurgery, and aims to create a first-in-class technology platform for personalized brain cancer care through 10-second delivery of accurate molecular pathology information to guide various clinical decision makings in the operating room and beyond, more efficiently than currently possible. By incorporating state-of-the art machine learning algorithms into a `true-and-tried' rapid diagnosis technology entitled "Picosecond InfraRed Laser Mass Spectrometry (PIRL-MS)" the team aims to extract detailed molecular pathology information through augmented spectral analysis focused on determining detailed prognostic molecular pathology markers of suitable surgical margin extents, aggressiveness of resection and response to various adjuvant therapies. Currently, the presence of these markers is often confirmed days post-surgery, resulting in potential suboptimal treatment planning that may require revision surgeries or delayed planning of post-surgical treatments. Building on published success of PIRL-MS in 10-second determination of many morphometrically or molecularly distinct types/subtypes of human cancers, this proposal aims to finetune existing data analysis workflows through augmentation with artificial intelligence to extract additional molecular pathology information from PIRL-MS spectra of human brain cancers. Here, the previously demonstrated sensitivity of PIRL-MS signal to various metabolic alterations that take place within cancerous tissues in response to the presence of several diagnostic/prognostic markers such as those associated with cancer `driver mutations' in critical RAF kinase proteins (e.g. BRAF), or in isocitrate dehydrogenase-1 further de-risks the proposed research.Upon securing institutional authorization for a retrospective human tissue study, we seek to exhaustively sample over 1,200 banked adult brain cancer tissues with PIRL-MS to catalogue spectral profiles, and by incorporating `supervised analysis' using various multivariate methods augmented with machine learning approaches including random forest (RF), support vector machines (SVM) and convolutional neuronal networks (CNN) analyses, our proposal aims to extract PIRL-MS spectral features that are statistically correlated with the presence of various molecular pathology prognostic markers (verified against pathology records).

The timely supply of tools and functional/repair parts, currently limited by rockets, is critical for astronauts to solve unpredicted problems in space missions. The project explores the fabrication of tools/parts in-situ and on-demand by Laser Metal Deposition (LMD) Additive Manufacturing (AM) in zerogravity and vacuum. AM makes ready-to-use tools/parts directly from computer-aid models and feedstock materials either carried from Earth or recycled from space debris. Spacecraft printed in space can be thinner and larger, optimized for space environment, without the need to fold into rockets and survive vibrations at launch. However, current AM technologies on Earth will be obsolete in space due to distinctly different behaviors of materials in zerogravity and vacuum, e.g., dangerous powder floating in powder-based LMD and abnormal flow of molten metals.

The project explores wire-feed 3D/4D LMD AM for in space. First, we will investigate solid-fluid phase change in LMD AM in zerogravity and vacuum by numerical simulations due to limited access to space. A 3D transient model with volume of fluid free surface tracking will be used to probe the unstable thermocapillary convection of liquid bridge and molten metal pool in LMD by considering mass, heat, and momentum conservations with parameters relating to metal addition, surface tension, phase change, Marangoni stress, material properties, heat transfer and laser heat input. Second, we will probe fundamental laws that govern 3D-to-4D shape-morphing of printed parts subjected to space stimuli by multiphysics finite element method. Special focus is the multi-material 4D printing with shape memory alloy. Third, we will validate the findings experimentally on Earth, where a robotic wire-feed LMD 3D printer with a robotic built table will print parts in different orientations in a vacuum chamber. Heat and mass transfer will be examined by high-speed cameras. Temperature and solidification parameters under different process parameters will be measured. Precision and quality of printed parts will be characterized at macro/micro scales. Machine learning will be used to reveal the complex combinations of structure-(physical parameters in space)-processing-properties-performance of LMD AM in space that could lead to proper designs and optimal process parameters.

The proposed development of in-space 3D and 4D LMD AM is high-risk. If successful, it will expand our ability in deep space exploration and colonization.

Industry 4.0 promises to usher the construction industry into a new era, where WALL-E like robotic pals will work seamlessly alongside human workers, collaborating on dexterous, physically demanding, and dangerous tasks. Although existing robots are designed to perform inherently repetitive and potentially dangerous tasks at a higher speed, power, and precision than their human counterpart; the unstructured and dynamic nature of construction sites relies on a collaborative and co-dependent situation where human workers assist or supervise robots with countless tasks. As more robots will be utilized onsite, holistic consideration for successful human-robot interaction (HRI) needs to be established. There are countless questions to be addressed regarding the robot's knowledge, awareness, and response to how human workers learn, move, feel, and make decisions onsite.

This research program aims to improve the working dynamics of the robot-human pair in distracting, complex, and busy construction sites by characterizing cognitive/emotional processes and behaviors in human workers to inform the design and control of human-centered construction robots. Because of the interdisciplinary nature of integrating human and robot factors to create automatized construction sites, we comprise a diverse team with complementary expertise in Civil Engineering, Cognitive Science, Robotics, and Biomedical Engineering. Our work will establish physiobiological markers and empirical measurements received by construction robots to improve their ability to understand workers' action goals, motor abilities, and social-emotional states. To achieve our goal, we aim to: characterize workers' cognitive abilities (e.g., attention), goal-directed actions, psychological states (e.g., anxiety), and perception (or misperception) of their robot co-workers using psychophysical and biometric recordings; and feed these measurements into the reinforcement learning framework for optimal construction robot control policies.

This work is ambitious because the mechanism of human cognitive/emotional process and behaviors is not fully understood, and its role in HRI at construction environments has not been examined. This work will have a significant impact on creating safe, human-friendly, and cost-effective construction sites and helping 1.4 million Canadian construction workers prepare for the new era of automated construction sites and collaborate with robot co-workers with greater confidence. 

According to governmental reports, the proportion of senior citizens in the Canadian population is expected to double by 2025. Voice assistant bots (e.g., Amazon Alexa) could support safe aging in a home environment, by assisting with medication reminders, offering guidance with challenging tasks, or even providing a form of companionship. However, seniors report challenges with using bots (e.g., forgetting to call upon the bot to activate it, struggling to get the bot to respond to their tone or accent). Creating bots that are adapted to the needs of seniors can bridge gaps in the effectiveness of this technology for this population. Leveraging interdisciplinary approaches from computer science, psychology, and cognitive neuroscience, the objective of this proposal is to uncover the types of personalized adaptations that will yield a greater uptake of assistant bots among seniors. The long-term impact of this project will be to develop new technology to facilitate seniors' independence in home settings.

We will investigate three forms of adaptations: 1) video-voice bots that prompt users based on their inferred need states (e.g., frustration, panic; based on physiological data and face videos), 2) bots that match the lexical structure (words and sentence structure used) and prosody (tone of voice, accent) of their users, and 3) bots whose video and voice personas are individualized to users' preferences. These bot adaptations have never been combined, nor tested with a population that is reticent to their use. In collaboration with Long Term Care facilities and community homes, we will determine how each form of adaptation impacts senior users' uptake and satisfaction with the bot, in difficult laboratory tasks and in daily activities at home. Attention will be given to individualizing the bot to users' diverse identities and cultural backgrounds. This interdisciplinary project requires expertise in computer science (to create novel bots) as well as cognitive neuroscience and psychology (to infer users' cognitive and emotional states based on physiological data and prosody), to address a public health challenge (facilitating senior citizens' use of technology to promote independent living). Developing new forms of technology that will be readily accepted by seniors is a high-risk endeavour. If successful, this project will help ease the load on an already overburdened health care system and reduce barriers to independence in a vulnerable population.