Award Recipients: 2025 Exploration

Neuroscientific research has established a fundamental distinction between detailed and schematic (gist-like) information processing that is conserved across species and across ontogenetic development. From an evolutionary perspective, these dual modes reflect adaptive trade-offs. During development, biases toward detail or schema shape trajectories of learning, remembering, thinking, and reasoning. Our central hypothesis is that individual differences in detailed vs. schematic processing form a common thread across domains: bias toward detail supports rich mnemonic representations, but increases vulnerability to trauma-related memory intrusions. Bias toward schemas supports resilience following trauma and enhances abstract reasoning, the latter being essential for success in science, technology, engineering and mathematical (STEM) professions.

In an interdisciplinary collaboration including evolutionary biology, experimental and systems neuroscience, developmental cognitive neuroscience, clinical psychology, and education, we will probe individual differences in detailed vs. schematic processing in flies, rodents, and humans. Specifically, we will examine responses to environmental stressors and the application of schematic reasoning in light of individual differences, especially biological sex, which moderates both resilience to trauma and STEM achievement.

In non-human organisms, the detailed vs. schematic processing modes will be operationalized using the classic WHAT (identity) vs. WHERE (location) distinction, providing a platform for probing neural circuits that are preserved across species from flies to humans. We will test manipulations of WHAT vs. WHERE circuits to determine the most effective methods for leveraging individualized detailed vs. schematic processes in relation to stress resilience and acquisition of schematic knowledge, stratified by biological sex. We will model selective tuning of WHAT and WHERE brain circuits—derived from non-human results—using fMRI data from males and females with varying responses to stressful events and STEM achievement.

Translation of non-human circuit manipulations into human neuroimaging models will provide a basis for neurally informed, precision interventions to promote psychological resilience and STEM achievement that are sensitive to sex. By linking evolution, mental health, and education, this research ushers in a transformative new paradigm for understanding cognition and emotion.

OBJECTIVES:

Physical activity and social engagement are key factors of healthy aging, yet many older adults lack access to programs that are physically stimulating and intrinsically motivating. Novel, interactive technologies leveraging art or music can make physical activity more enjoyable and engaging, particularly in groups lacking intrinsic motivation to do regular exercise. Thus, the objective is to develop wearable movement-to-art technology (WMAT) – a specialized garment with embedded sensors to detect body motion and convert it into real-time music or visual art. The aims are to: 1) establish feasibility and usability and validate a WMAT; 2) evaluate implementation in movement therapy sessions to improve physical function, motivation and quality of life in older adults; 3) showcase participants’ art in a public exhibition to foster community connection and reduce social isolation.

RESEARCH APPROACH:

Phase 1 – Technology development: Engineering and music experts will develop and validate the WMAT by engaging older adults throughout the design process to ensure comfort, acceptability, and controllability, in addition to ensuring technical capability.

Phase 2 – Therapeutic implementation: Participants will regularly engage in movement-based art sessions (choreographed or improvisational) highlighting the WMAT. Semi-structured interviews will assess perceived changes in quality of life, motivation, engagement and social connectedness; functional mobility and gait assessments will quantify physical benefits. Results from Phases 1-2 will be disseminated in academic journals and conferences.

Phase 3 – Knowledge translation: Participant music and artwork will be curated into a public exhibition celebrating their creative agency and emphasizing community engagement.

NOVELTY AND SIGNIFICANCE:

While music is commonly used to engage adults with exercise through dance, emerging technologies can directly convert bodily movement into creative artistic expression, with immense potential to increase enjoyment and adherence to regular exercise. Leveraging synergies in engineering, health sciences, social sciences and art will result in an impactful, evidence-informed WMAT with potential for diverse communities and clinical settings. Successful outcomes could expand therapeutic options beyond conventional exercise programs lacking creative expression, improving quality of life, encouraging sustained physical activity and fostering social connection in aging adults

DNA and RNA sequencing technologies have transformed biology and medicine, yet even the most advanced platforms face serious limitations. Existing methods often require multiple steps—such as amplification, reverse transcription, and chemical modification—that can introduce noise, bias, and errors. Detection systems based on fundamental particles such as ions or photons typically yield low signal-to-noise ratios and identifying non canonical nucleotides remains especially challenging. Together, these issues limit both the efficiency and precision of sequencing. We propose to use electrons as the detection particles in a fundamentally new approach to sequencing. By harnessing unidirectional electron transport in chiral nucleic acids, we enable ultrafast, background free detection that improves accuracy and efficiency, with the potential to transform genomics.

Transformative mechanisms: Unlike traditional methods that rely on ions to generate current or photons for fluorescence detection, our approach uses fundamental particle of electrons, which are photogenerated by an ultrafast laser. The unique chiral structure of RNA enables us to exploit the chirality-induced spin selectivity (CISS) effect, directing electrons to move unidirectionally along the RNA strand. By directly detecting ultrafast photocurrent, this dramatically boosts the signal-to-noise ratio, enabling ultra-precise detection at unprecedented speeds.

Research tasks: First, we will engineer DNA/RNA nanostructured photoconductive devices in which nucleotide strands functionalized with metal nanoparticles self-assemble on gold electrodes. We will then employ advanced sub-picosecond ultrafast photocurrent spectroscopy to track electron dynamics under a range of conditions (pH levels, temperatures, and magnetic fields) informing optimal sequencing conditions. Finally, we will apply these insights to distinguish native RNA from chemically modified variants, enabling direct, accurate readouts and opening a new frontier in molecular analysis.

Novelty and significance: By integrating expertise in advanced optoelectronics and molecular biology, our high-risk, high-reward project pioneers a fundamentally new approach to nucleic acid sequencing, one that could redefine the fields of genomics and transcriptomics. Modern biology relies heavily on these technologies, and our advancements will position Canada at the forefront of global innovation in DNA/RNA sequencing, genetics, and biomedical research.

Despite significant advancements in health and technology, deep-seated inequities persist in Canada's Indigenous and Black communities. These disparities stem from historical injustices and ongoing systemic issues, including colonialism and racism embedded in data, research paradigms and the marginalization of traditional knowledge. Conventional research, and increasingly, AI applications, can perpetuate these harms by ignoring unique cultural contexts, data biases, and lived experiences of affected communities. There is an urgent need to rethink how technology can serve, not harm, underrepresented populations. This proposal outlines co-creating a Decolonized and Anti-Racist AI framework to identify, understand and address these disparities. Our approach combines cutting-edge AI (natural language processing, machine learning, and responsible AI) with Indigenous knowledge, decolonial methods, and critical social dimensions in health.

Led by a diverse team including Indigenous and Black scholars, alongside expertise in health equity, this project moves beyond simple “fixes” to develop foundational principles for AI that truly serves these populations. This is not about applying AI to existing problems but fundamentally re-envisioning how AI is developed and deployed in a culturally safe, ethical, and empowering way.

Our goal is to challenge existing norms by designing and testing a framework where AI is developed with Indigenous and Black communities, ensuring cultural safety, data sovereignty, and a commitment to anti-racism and decolonization. This involves bridging different ways of knowing, challenging traditional power dynamics in research, and navigating complex ethical landscapes with no established precedents.

Ultimately, this project aims to foster community empowerment, create equitable technological solutions, and establish a new model for responsible, interdisciplinary research that directly benefits marginalized groups. We will:

1. Co-Design AI from the Ground Up: Develop AI models and data protocols with Indigenous and Black community partners, challenging traditional top-down research.

2. Integrate Decolonial and Anti-Racist Principles: Embed these principles directly into AI design.

3. Explore AI for Systemic Advocacy: Use AI to champion health and social justice, community-led advocacy and resource mobilization.

Success promises a major impact on health equity, pioneering a new model for responsible technological innovation.

This research challenges the exclusionary paradox in contemporary architectural practice, as participatory design methods remain performative and fail to transfer real agency to communities. We posit that true co-design requires an equitable sharing of design tools, specifically, digital and computational tools that remain inaccessible to racialized communities, and often reproduce colonial spatial logics.

This interdisciplinary collaboration is between scholars in the fields of Architecture and Information Technology, in addition to partnerships with community care workers. The partnership is built on a proof-of-concept workshop, where the PIs presented a digital simulation of the Russell Heights community, which had garnered community support and requests for further digital engagement tools.

Therefore, we are proposing a framework that merges anti-racist architectural praxis with game-based co-design methods.

This research has three main objectives: First, to experimentally explore play-based co-design tools that democratise access to design resources, especially geared at racialized youth; second, to analyse, pinpoint, and counter the colonial, extractive methods in architecture and urban design as reproduced by extractive and superficial community consultations. Third, to build a living repository of architecture innovation that facilitates design for non-architects.

Working with two predominantly Black communities in Ottawa, the Russell Heights community and the Strathcona Heights community, the methodology employs research-creation by using action research and research by design methods to design and test digital toolkits through iterative cycles of community fieldwork. These include bridging play with digital architectural platforms and game-based educational tools that demystify zoning codes through interactive tools. This approach fundamentally rethinks architectural tools not as professional monopolies but as shared community resources, a necessary condition for moving beyond performative participation, and toward co-production of the city.

This research takes on the risk of moving two highly commercialized fields (architecture/gaming) into the realm of grassroots activism, which requires rigorous social and technical research. In return, the project aims for the high reward of a generalizable framework for democratizing architecture and simulation technology, disrupting top-down design methods, and centring community agency in architecture.

En 2023, un tiers des personnes adultes Canadiennes, soit plus de 11 millions de personnes, étaient perçues comme obèses (pcO) selon l’indice de masse corporelle (IMC). La réalité de l’obésité dépasse pourtant largement l’IMC. Un consensus publié en 2025 dans The Lancet reconnait l’obésité comme une maladie chronique complexe et appelle à une évaluation individualisée centrée sur les impacts fonctionnels. Malgré cela, une personne sur trois pcO continue de subir de la grossophobie médicale. Cette forme de discrimination systémique, souvent méconnue de ces personnes et même taboue dans le milieu de la santé, entraîne des conséquences graves : stigmatisation, minimisation des symptômes et biais de prise en charge qui nuisent directement à la qualité des soins. Le projet vise un changement de paradigme face à la grossophobie médicale par une démarche de recherche créative-action-participative, transdisciplinaire, intersectorielle et équitable. Il s’appuie sur deux piliers fondamentaux : (1) une équipe diversifiée composée de personnes pcO, militantes, artistes, soignantes, intervenantes communautaires, décideuses, et chercheuses et (2) un médium puissant mais encore sous-utilisé dans le domaine médical pour déconstruire les stéréotypes: la bande dessinée (BD). En s’inspirant d’expériences vécues recueillies lors d’entrevues inclusives et d’ateliers créatifs, une BD inclusive et accessible sera cocréée en français et en anglais. Elle représentera des corps variés et sera enrichie d’un dictionnaire visuel. L’analyse comparative fondée sur le sexe et le genre permettra d'explorer les discriminations croisées. Les organismes communautaires partenaires, experts en accessibilité et milieux spécialisés assureront la participation pleine et entière des personnes pcO, notamment celles issues de communautés historiquement marginalisées. La sécurisation culturelle et psychologique sera intégrée à chaque étape. La BD sera utilisée comme outil de formation pour les personnes soignantes en milieu universitaire et professionnel, et de dialogue renouvelé entre les personnes pcO et soignantes. Les impacts de cette démarche sur les biais implicites et le pouvoir décisionnel des personnes pcO et soignantes seront confrontés. Par cette approche art-science novatrice, le projet vise à réduire la grossophobie médicale systémique, transformer la relation de soins et améliorer les pratiques cliniques, pédagogiques et de recherche afin de renforcer la santé globale des personnes pcO.

Context-In 2024, Jeu-Aide et Référence (JAR), Quebec’s provincial help line for people experiencing gambling-related problems, recorded a 21% increase in calls. Driving this increase is the fact that gambling games are now accessible 24/7 on our mobile devices. This situation enables gambling operators to personalize their offer. They gather granular data on each player; with this, they can make detailed player profiles, which can be used to ‘nudge’ players to keep playing. JAR has developed TeleCounseling interventions devoted to helping people who struggle with gambling problems. Not covered in the counseling repertoire, however, are the ways players are profiled, targeted and manipulated in the age of addiction by design.

Objective-At Concordia University, the new CHANCE Collaboratory is developing tools to teach people about addiction by design. This NFRF-E project builds on that platform. Placing the front-line work of JAR at its centre, this project’s main objective is to address JAR’s knowledge needs.

Research Questions-(1) Are there opportunities for addressing addiction by design within JAR’s TeleCounseling workflow? (2) Of the types of mobile games most mentioned by callers to JAR, what do their terms of use and privacy policies permit, and how do these apps actually behave in practice? (3) Of the harm reduction apps that JAR recommends to clients, what are their terms of use and privacy policies, and how do they actually behave in practice? (4) (How) do these apps interact?

Methods-(1) Focus groups with JAR Directors and Counselors; (2) Walkthrough and Dynamic Analysis of Gambling Apps and Harm Reduction Apps.

Interdisciplinarity-We are social scientists, engineers and counselors/social workers.

EDI-Members of a community-advisory board will be purposively selected to ensure representation of maximum diversity within our target population. Our approach is fundamentally intersectional.

High risk—There are technical risks related to constraints imposed by powerful platforms (e.g. Google).

High Reward—1) we create interventions that helps people struggling with gaming- and gambling-related harms; 2) we help to shift the field of gaming- and gambling-harm prevention further towards interventions that address underlying, manipulative capacities of operators.

Feasibility—we build on extant relationships, and our laboratory infrastructure at Concordia University.

Delays in diagnosing common infections (urinary tract infections, respiratory illnesses) and conditions such as dry eye disease impose substantial burdens on patients and healthcare systems. Standard laboratory tests typically take a few days and cost about tens to hundreds of dollars per assay. This project will develop a low‑cost, label‑free diagnostic platform that analyzes the behavior of small biofluid droplets (urine, saliva/sputum, tears, capillary blood) as they impact, spread, evaporate, and freeze on antibacterial surfaces fabricated via the thermal‑spray process. Tunable Cu/Ag‑coated surfaces will be produced to optimize droplet interaction while providing inherent antimicrobial activity. Beyond impact dynamics, as each droplet dries or freezes on these functional coatings, distinctive features emerge--maximum spreading diameter, peripheral ring deposits (coffee ring), and frozen droplet shape. These signatures encode changes in viscosity, surface tension, and biochemical composition associated with disease; for example, respiratory infections increase saliva/sputum viscoelasticity, reducing spread and altering rebound, while dry‑eye tears disrupt normal ferning patterns. The workflow uses minimally invasive sampling (clean‑catch urine, expectorated saliva/sputum, microlitre tears, or finger‑prick blood), deposition onto a coated coupon, and smartphone imaging of impact, drying, and freezing. Numerical simulations of droplet impact with phase change and biological modeling of microbial growth (e.g., biosurfactant‑driven reductions in surface tension) will guide features extraction and expand experiential dataset. Machine‑learning models trained on the resulting image/video dataset will classify target conditions (e.g., UTI‑positive urine, dry‑eye tear imbalance, respiratory infection from saliva/sputum). A 3D‑printed prototype–integrating illumination on a compact tripod, a replaceable coated‑coupon holder, and a smartphone imaging mount–will be developed. Fabrication, coating deposition, and characterization will be performed using state‑of‑the‑art thermal spray and materials characterization laboratories. By integrating surface engineering, fluid mechanics, rheology, biological modeling, and AI, the project aims to deliver a reagent‑free, rapid, and accessible diagnostic system for remote, underserved, and surge‑capacity care.

This project develops the first culturally-adapted generative AI system for mental health interventions with forcibly displaced persons experiencing trauma. Nigeria's displacement crisis affects over 3.2 million people, with mental health disorder prevalence of 75-94% for PTSD, 60-98% for depression, and 66-99% for anxiety. With fewer than 300 psychiatrists serving 200+ million people, conventional approaches cannot scale to meet demand. An interdisciplinary approach involving experts in AI, global health, and cultural anthropology working together is essential. While AI-health collaborations are emerging, collaborations with cultural anthropologists remain rare despite their transformative potential for creating culturally-grounded technologies.

OBJECTIVES: We will: (1) create AI models incorporating local expressions of distress and healing; (2) implement participatory design to co-create systems with affected communities; (3) establish human-AI collaboration frameworks for humanitarian settings.

APPROACH: Using participatory design, we will work with displaced populations to encode culturally specific trauma expressions into lightweight AI models. Through community storytelling sessions and focus groups, we will systematically catalog local terminology for distress, indigenous coping strategies, and healing narratives. These cultural expressions will inform AI training to recognize and respond using locally relevant language. We will implement safety protocols and pilot-test with 30-50 participants across multiple displacement sites.

HIGH-RISK INNOVATION: This radically challenges current AI development paradigms by inverting the standard Western-centric model creation process. We develop novel methods centering indigenous knowledge from inception, requiring unprecedented integration of rarely combined ethnographic, computational, and clinical disciplines. We pioneer new techniques for encoding cultural trauma expressions, fundamentally challenging how machine learning systems are conceptualized.

TRANSFORMATIVE POTENTIAL: Success would establish new paradigms for ethical AI development globally, with frameworks applicable to other vulnerable populations, refugee contexts, and indigenous communities worldwide. This would demonstrate scalable solutions for 100+ million displaced people and open new interdisciplinary research areas at the intersection of cultural preservation and technological innovation.

Artificial Intelligence (AI) is increasingly shaping how young people access information about sexuality. Global guidelines on navigating AI for children and youth (herein youth) are primarily risk-focused and are not inclusive of disabled youth’s unique needs, which ignores how AI can potentially enhance safety, agency, and inclusion around sexuality for the 1.3 million Canadians aged 15-29 years who report living with a disability and/or chronic illness. Disability disproportionality impacts youth from structurally marginalised Canadian communities.

This interdisciplinary, youth-centred project challenges entrenched narratives framing disability and sexuality solely as abuse prevention, risk reduction and the management of behaviours deemed ‘inappropriate’ by society. While safeguarding is critical, overemphasis on risk has suppressed exploration of how AI can support disabled youth’s sexual rights, identity, agency and autonomy. Our approach brings together the historically siloed fields of disability studies, sexuality education, digitalisation, human computer engineering and AI ethics to develop the first Disability Inclusive AI Sexuality Toolkit to guide developers, educators, and policymakers in creating accessible, affirming AI-enabled resources on sexuality topics.

Over two years, our objectives are: (1) To map global AI and youth safety guidelines to systematically identify gaps in disability sexuality inclusion; (2) To convene disabled youth, designers, engineers, ethicists, and HCI experts in co-creation workshops to develop inclusive AI principles and protocols; and (3) To produce and pilot a globally relevant, accessible toolkit. This work will disrupt risk-only narratives, reframing AI as a potential enabler of sexual rights for disabled youth.

This work is high risk as it addresses a rapidly evolving, yet contentious field of AI digitalisation for disabled youth and combines disciplines from diverse epistemologies to disrupt protectionist norms in disability. It has the potential to bring transformative rewards by creating the first national guidelines with global significance that integrates sexual rights, accessibility and inclusive AI design with a youth disability focus. It can also inform policy and national accessibility design standards that reimagine AI as accessible, affordable and affirming of disabled identities rather than solely risk reduction.

Mechanochemistry, the induction of chemical transformations by mechanical force in the solid-state has emerged as a powerful solvent-free approach to chemical synthesis. Solvent-free chemical processes have been recognized as one of the top ten innovations to make our planet more sustainable by the International Union of Pure and Applied Chemistry (IUPAC), reflecting their potential for transformative impact in science and industry. However, the use of mechanochemical techniques to manipulate supramolecular phenomena involving ‘biologics’ (e.g., nucleic acids, polypeptides, etc.) remains completely unexplored. Such phenomena include e.g., the (dis)assembly, (dis)aggregation, and (un)folding of biologics, which are critical to their performance. These phenomena are traditionally manipulated via solution processes that generate considerable waste. Moreover, the desirable solid-state formulation of biologics for long-term storage or transport often leads to unpredictable damage during drying. Hence, understanding how to manipulate biologics in the solid-state is thus fundamentally important for reducing waste associated with manufacturing processes and for producing stable formulations with predictable properties across a wide range of applications.

Project RE-FORM delves into a fundamental study of the mechanochemical stimulation of supramolecular phenomena of importance for biologics, to explore greener and more efficient ways of manipulating, stabilizing, and formulating biologics for biomedical research. Because of the pioneering nature of this project and the consequent risk, a large parameter space must be mapped experimentally. We will therefore implement a high-throughput screening approach using relevant model systems of increasing complexity. These will include models for exploring nucleic acid hybridization, peptide aggregation, and nucleic acid/protein folding. These model systems are modular, which will provide fundamental tools to investigate the energetics and kinetics of the biologics under different solid-state mechanical stresses. Exploration of this parameter space will be guided by in silico tools, on one hand to guide conditions towards a desired goal, and on another to dissect complex trends and elucidate underlying biophysical mechanisms.

This project aims to transform environmental health monitoring by creating a real-time detection and rapid intervention system for allergenic pollen exposure. The goal is to develop an automated sensing network capable of identifying, quantifying, and reporting airborne pollen species with high precision and minimal delay, enabling timely protective measures for individuals and supporting public health responses.

The system will combine advanced robotics, computer vision, optical imaging, and artificial intelligence. Airborne particles will be concentrated onto collection tapes and imaged by robotic microscopes using Fourier-based super-resolution techniques to overcome the limits of conventional optics. Machine learning models will then detect, count, and classify pollen grains, determining species origin and allergenic potential. Environmental parameters such as particulate matter, humidity, temperature, and air pressure will also be recorded to provide context for the data.

This highly interdisciplinary approach integrates environmental science for accurate pollen identification, healthcare expertise to link pollen levels with allergy-related health events, and psychology to design an AI-driven communication module. Drawing on social psychology research, this module will transform and translate the collected data into clear, actionable health messages that encourage users to take protective steps when exposure risk is high, ultimately improving their health outcomes. A mobile application will let users log symptoms and receive personalized alerts, while ensuring their personal data remains private.

The novelty lies in uniting autonomous environmental sampling, super-resolution imaging, and adaptive machine learning into a single, fully integrated platform for real-time allergen detection. If successful, the system will not only transform allergy management but also serve as a foundation for broader biological and environmental monitoring applications, including the detection of airborne pathogens and harmful pollutants. By enabling faster, more accurate detection of environmental health threats, this innovation has the potential to enhance quality of life, inform public health policy, and strengthen community resilience against environmental and biological risks.

This research offers an innovative and interdisciplinary approach to mapping neuronal excitability by integrating principles from neurology and quantum physics, with epilepsy as the target neurological disorder.

Neuronal excitability refers to the brain's responsiveness to stimuli and is dynamically modulated across regions. Traditionally, excitability is quantified by responses to external stimuli, such as transcranial magnetic stimulation. However, limitations in these methods make whole-brain excitability mapping impossible. Yet, excitability is a fundamental property, altered in nearly all neuropsychiatric conditions, and its measurement could significantly aid in diagnosis, prognosis, and treatment guidance. Hyperexcitability is the cornerstone of epilepsy, a neurological condition characterized by recurrent seizures affecting over 200,000 Canadians. Since no direct excitability assessment is performed, treatment relies on the trial-and-error administration of medications that inhibit excitability, often leaving patients with unresolved seizures and social exclusion for years.

For the first time, we propose to map excitability by applying quantum mechanics principles and leveraging resting-state electrical (EEG) or hemodynamic (fMRI) brain activity. We hypothesize that spontaneous activity reflects the properties of neural circuits, including their activation thresholds and responses, and thus reflects neuronal excitability even in the absence of explicit stimuli. We conceptualize brain activity as a 'brain state' or probabilistic quantum state, with a likelihood of change over time. A higher probability of state change indicates a system more primed to respond to stimuli, or more excitable. This approach allows us to measure excitability as a probability function at any brain point, at any time, enabling whole brain mapping. Our goals are to (a) develop quantum excitability maps from EEG and fMRI data, (b) test their sensitivity to hyperexcitability by comparing epilepsy patients with controls and assessing medication effects within epilepsy patients, and (c) evaluate their potential as predictive biomarkers for responses to antiseizure medications. This research will benefit patients with epilepsy, enhance our understanding of brain physiology, and allow the study of excitability dysfunctions in Alzheimer's, ALS, and disorders of consciousness. Additionally, it will create an interdisciplinary, diverse, and enriching training environment.

Canada is experiencing a sharp rise in hate crimes: 4,777 police-reported cases in 2023, a 145% increase since 2019. These acts disproportionately target people based on race, religion, sexual orientation, language, and age. Addressing this pressing issue requires bold, interdisciplinary innovation.

This project seeks to find if and how immersive learning (IL) through virtual reality (VR) can strengthen mutual understanding and foster empathy, focusing on racially and culturally minoritized communities. These communities share experiences of displacement, marginalization, cultural preservation under pressure, and recovery from trauma, forming the foundation of our AI-powered VR empathy scenarios (AVS).

Each AVS features autonomous virtual agents (AVA) whose behaviours are shaped by policies trained on early-phase interaction data and refined via a machine learning technique called reinforcement learning from human feedback. AVA respond dynamically to user behaviour, creating a living, responsive environment that deepens emotional engagement and empathy.

Objectives

1. Co-create AVS with multiple racially and culturally minoritized communities.

2. Evaluate if experiencing one community’s AVS enhances empathy toward another.

3. Mobilize AVS for broader application across Canada.

Research Questions

1. How does exposure to one community’s AVS affect empathy toward another?

2. How do users emotionally and cognitively respond to AVS based on lived experiences of marginalization?

3. How does the adaptive behaviour of AVA shape empathy development across user groups?

Methodology

This two-year project will be hosted at Edmonton and Calgary public libraries, accessible, inclusive spaces ideal for community engagement. It will proceed in three phases:

1. Participatory scenario development with community members.

2. VR production and AI integration, refined by community feedback.

3. Implementation and evaluation via surveys, interviews, and behavioural observation.

Outcomes

This high-risk, high-reward initiative combines AI, immersive learning, and participatory design to address a pressing social challenge. Its interdisciplinary scope, spanning education, design, computer science, psychology, and community engagement, positions it to generate scalable, mobile solutions. Grounded in equity, diversity, and inclusion, and leveraging the reach of public libraries, the project offers a feasible and replicable model to foster empathy and resilience across Canada.

Peripheral nerve (PN) injury can lead to major disabilities such as paralysis of a limb. Although injured PN regenerate spontaneously, the regeneration process is often slow and not complete, compromising the likelihood of motor recovery. Moreover, surgical interventions for nerve repair often yield poor outcomes. Both intraoperative and brief postoperative electrical stimulation (ES) of the damaged nerve site has been demonstrated to accelerate axonal regeneration. Current technologies for ES such as wired electrodes are limited to intraoperative use only since they require a wired connection to bulky stimulator and often surgeries to remove them. Here we propose to design, develop and validate a wireless electrical PN stimulation implant that is small, biocompatible and bioresorbable and can be placed around the proximal stump of the injured nerve site during surgery. The implant will be based on soft piezoelectric thin film nanogenerator which can be excited wirelessly by programmable ultrasonic (US) stimulation, removing the need for a battery in the implant. A soft and flexible US active nanogenerator from Poly(l-lactic acid) (PLLA) based nanofibers with excellent piezoelectric properties will be developed. Molybdenum (Mo) electrodes will be deposited on the PLLA film to provide the ES. The implant will be placed around the proximal nerve stump and will dissolve in biofluids after the brief stimulation period. After surgery the implant can be driven wirelessly by placing an US transducer outside on the skin. Initially ES implant output at various distances from the US transducer in different media will be measured. Implant biocompatibility, operational stability and bioresorbability will be determined in vitro. To test the benefit of nerve ES in vivo, a lesion of 15 mm in the sciatic nerve of Lewis rats will be removed. The proximal nerve stump will be wrapped (or not) with an implant producing (or not) ES. Improvement of nerve regeneration by ES will be confirmed by quantification of the number of nerve fibers and distance regenerated 4 weeks after surgery. With the ability to provide wireless ES (without a battery on the implant) and dissolve in biofluid the proposed novel implant can overcome limitations of current ES technologies. It will open the door for providing postoperative ES without a removal surgery resulting in hopefully significant improvement in PN regeneration and patient recovery.

Contemporary society’s dependence on fossil fuels is recognized as the main contributor to the climate change crisis, and climate plans that do not involve divestment from the exploitation of oil, gas and coal are more and more recognized as inadequate to reach carbon neutrality. While renewable energy sources – especially wind and solar energy – are becoming increasingly cheap, the intermittence of their production make them ill-suited to many applications that require energy on demand. As a result, energy storage technologies, because they enable the collection of energy in times of high production and its release in times of peak demand are a key component of a green energetic transition, especially for remote communities.

This project will explore pure metals as energy vectors. This idea will attempt to exploit the fact that earth-abundant metals such as aluminum and iron possess a high energy density that can conceptually be released by oxidation (combustion). However, while the resulting metal oxides could be regenerated to the pure metal by smelting, traditional methods are energy-costly and have a high carbon footprint. To address these issues, our proposal brings together an organometallic chemist, an electrochemist and an engineer to design a fully circular model for metals as energy vectors. Using specially designed N-heterocyclic carbenes (NHCs) – highly reactive organic ligands – we will solubilize metal oxides to enable their recycling through electrowinning – an energy-efficient method that can use electricity from renewable sources. The interaction between combustion conditions and the efficiency of NHC solubilization and electrowinning efficiency will be explored. Our proposal is:

High risk: our interdisciplinary team will look at all the stages of beyond-state¬-of-the-art energy storage. This fundamental reinvention of metal solubilization will be coupled their implementation as fully circular energy.

High reward: we will be working on unexplored chemistries for one of the most important technological challenges of our era in terms of energy and climate: a fully cyclable energy vector that will allow the on-demand storage of electric energy that is suitable for release on-demand.

Interdisciplinary: we are a diverse, team of experts from fields that do interact too seldom: organometallic chemistry, electrochemistry and mechanical engineering. Our team has the expertise to bring this project from initial design to scalable deliverables.

From Elite Labs to Every Desk: Democratizing Lab-on-a-Chip technologies through knowledge dissemination and community building

Summary: We propose an initiative to transform lab-on-a-chip (LoC) manufacturing—shifting from a specialized laboratory process to a globally accessible platform that enables rapid, distributed innovation in research and diagnostics. By harnessing low-cost 3D printing and accessible digital manufacturing tools, we will test the feasibility of LoC production by diverse participants worldwide and build an international framework to accelerate innovation and empowering local diagnostic capabilities.

Aim 1: Global network for remote, distributed digital manufacturing

We will recruit and coordinate an international cohort of participants, spanning academic labs, small enterprises and citizen scientists, to fabricate LoC devices using accessible manufacturing tools. This network will serve as a real-world testbed for evaluating the scalability and inclusivity of distributed manufacturing.

Aim 2: Standardization and benchmarking

We will develop and disseminate open-source standardized fabrication protocols, material kits, reference reagents, and quantitative performance metrics. These resources will form the foundation for global standard in LoC research and application.

Aim 3: Feasibility, knowledge dissemination and societal impact

We will assess the technical feasibility of remote LoC fabrication, measure the extent and depth of knowledge transfer, and examine societal and regulatory considerations. A particular focus will be on the implications of locally manufactured diagnostic devices during public health crises, providing a blueprint for pandemic preparedness.

High-risk/high-reward: This proposal is high-risk because it probes a complementary alternative to the centralized model, testing whether a heterogeneous, globally dispersed network with variable local manufacturing conditions can reliably produce advanced biomedical devices. It is high-reward because success would democratize a transformative manufacturing technology, enabling communities worldwide to design, produce and deploy LoC diagnostics and other tools on demand. Such capability could revolutionize public health responses, accelerate scientific discovery, and establish a new paradigm for equitable access to advanced technologies. If successful, this work will redefine who can innovate, where innovation occurs, and how rapidly we can respond to global challenges.

Myelin—the insulating sheath around axons—is essential for fast, efficient neural communication. In diseases like multiple sclerosis, myelin loss leads to neurodegeneration and severe disability. Although some remyelination can occur, it is often inefficient or fails entirely, especially with age. No approved therapies directly target oligodendrocytes—the cells that produce myelin—to restore lost insulation. Developing strategies to enhance remyelination remains a critical unmet need.

Emerging research suggests two routes to myelin repair: one via oligodendrocyte precursor cells (OPCs) that mature into new myelin-producing cells, and another involving mature oligodendrocytes extending new sheaths. Even more intriguingly, mature myelin itself may exhibit structural plasticity, offering untapped repair potential. However, the signals driving these processes are still poorly understood.

A growing body of evidence suggests electrical activity plays a key role. Patterns of axonal firing influence OPC differentiation, sheath formation, and internode length—yet whether electrical activity alone, independent of biochemical signals, can drive myelination is unknown.

This project brings together a neural biologist, a materials scientist, and an electrical engineer to develop the first in vitro model of activity-driven myelination on synthetic axon-like fibers. These nanofibers will be fabricated via electrospinning, allowing precise control of diameter and alignment to mimic axonal structure and directionality.

A major innovation is our focus on ionically conductive fibers that better replicate the ionic signaling used by neurons—unlike traditional electron-conducting materials. Starting with electronic or mixed-conduction systems, we will engineer gel-like, ionically conductive fibers where ion concentrations can be tuned at the microscale.

Additionally, we will integrate surface modifications and drug-releasing coatings—via mesoporous structures or coaxial electrospinning—to study how axon-mimetic biochemical cues enhance OPC engagement and myelination.

This project is undoubtedly high-risk, high-reward. Recreating a functional, ionically conductive, synthetic model of myelination is very challenging, and has never been done. But success would transform our understanding of CNS repair and create a novel platform to identify therapies that promote remyelination—advancing both basic science and clinical translation.

Pesticides are essential in agriculture to reduce pest infestation and help attain required crop yields. Conventional spraying or soil application of pesticides results in only ~10% reaching target plants, and the remainder is lost to the environment. Many synthetic pesticides cause adverse effects to non-target organisms and damage the environment due to water contamination, biodiversity loss and degradation of arable lands. The objective of the research is to revolutionize pesticide application using innovative nanobiotechnology-inspired products and processes. We will demonstrate for the first time that the damaging effects of traditional use of synthetic pesticides can be cost-effectively eliminated by foliar application of biopesticides (plant-based chemicals, e.g., essential oils). Currently, essential oils are extracted from plants which is prohibitively costly. We will use essential oils (bioactive terpenoid compounds) that can be produced inexpensively in bioreactors by precisely engineered microbes, and enable their efficient uptake in plants by encapsulating them in biocompatible silica nanoparticles that penetrate leaf pores and migrate to tissues where pathogens proliferate, providing controlled release of the terpenoids. We will verify that the nano-encapsulated biopesticides do not pose unacceptable health risks to farmers and consumers, and that they are economically attractive and practical to apply. The efficacy of the nano-encapsulated biopesticides in protecting against an important bacterial pathogen that attack tomato plants will be assessed using advanced genomic assays. This project will thus provide new knowledge and novel methods for currently unsolved limitations in agriculture: (i) synthesis of nano-encapsulated biopesticides that are composed of environmentally benign materials; (ii) their crop application strategies that yield improved deterrence of plant pathogens of critical concern; (iii) demonstration of their safety to farm workers compared to traditional pesticides based on human cell line and in vivo murine toxicology assays; and (iv) evaluation of their positive impacts of adoption on farmers' livelihoods and market access, and how any barriers to their adoption can be overcome. The investigators will collaborate to demonstrate a paradigm-shift in pesticide application which will help cut pesticide use while ensuring food security for a growing global population.

There is a growing interest in personalized medicine, which aims to customize treatments based on an individual patient's unique health and behavioral profile. However, this approach depends on the availability of accurate and frequent personalized data. Although various wearable health sensors are available for continuous monitoring, most are limited to tracking biosignals such as heart rate, temperature, or blood pressure. Except for continuous glucose monitors, which monitors glucose in the skin interstitial fluid using microneedles electrodes, there are no devices that can track biomarkers, linked to health and disease, in vivo and in real time. Particularly, there are no real time health monitors that can analyze the most important class of biomarkers, proteins, due to the technical and scientific challenges related to continuously transducing signals in response to changing protein levels with sufficient sensitivity and specificity.

This proposal aims at developing the scientific breakthroughs that are needed for creating the next generation of wearable health sensors that can make personalized medicine a reality through the following aims. Aim 1 will focus on developing sensing surfaces that integrate aptamers and antibodies with electrochemical readout for specifically binding and analyzing a wide range of biomarkers, particularly proteins, in real time. Aim 2 is focused on sensors that leverage electrical stimulation in vivo to detect biomarkers such as proteins that have a low physiological concentration (e.g. pM in the skin interstitial fluid). Aim 3 is focused on testing the next generation of wearable sensors in vivo using relevant disease models including those representing acute conditions such as cardiovascular disease or chronic conditions such as diabetes.

This project is expected to develop the first wearable health sensor capable of detecting proteins at clinically relevant levels, which is critically needed for advancing personalized medicine. This work is anticipated to enhance healthcare quality, equity, and efficiency both in Canada and globally, with its commercialization offering the potential to generate significant intellectual property and economic benefits for Canada.

Neurological disorders impose an annual economic burden of nearly $800 billion in the United States alone (Gooch et al., 2017). Yet drug discovery productivity has declined, with an 8.4% drop in innovation for every billion dollars spent since 1950 (Scannell et al., 2012). This project seeks to overcome this bottleneck by creating an integrated ecosystem to accelerate antipsychotic drug discovery and delivery.

Molecular dynamics (MD) simulations are powerful tools for screening drug candidates, but remain computationally intensive and limited by current platforms. In parallel, preclinical testing often relies on behavioral or coarse physiological measures that provide only indirect insights into drug effects on the brain. These challenges underscore the urgent need for a more integrated approach.

The novelty lies in uniting two fields that have advanced largely in parallel: computational drug discovery and neurotechnology.

Massively parallel, reconfigurable hardware will accelerate MD simulations that now take weeks per compound. Deep learning (DL) will reduce the immense search space and optimize parameters, drastically shortening discovery cycles. Simultaneously, brain-computer interfaces (BCIs) will capture high-resolution neural data from animal models, decoded through custom hardware-software co-design for brain analytics. Integrating molecular simulations with neuroanalytics will enable unprecedented evaluation of drug efficacy directly at the neural network level, moving beyond traditional "black-box" behavioral endpoints.

The impact is twofold. Scientifically, this approach shifts drug testing from receptor-level analysis to system-level understanding of brain-drug interactions, improving targeting, reducing adverse effects, and lowering development costs. Societally, it offers the potential to transform treatment of psychotic and neurological disorders through more precise and effective pharmacotherapies.

This project is high-risk because it challenges the established "black-box" paradigm of neuropharmacology. Integrating custom hardware, DL-guided simulations, and the adoption of a new analytic framework has never been demonstrated, and barriers remain in performance, reproducibility, and workflow compatibility. Yet it is also high-reward: success would shorten discovery timelines, reduce costs, and improve outcomes by tailoring therapies to brain-network responses. This ecosystem could fundamentally transform pharmaceutical development.

Autism is a neurodevelopmental condition that affects social interaction, communication, sensorimotor processing and behaviour. Although it is often diagnosed in childhood, many individuals remain under-recognized and undiagnosed until adulthood, leading to significant challenges in receiving appropriate support and intervention. Current diagnostic methods for autism in adults are time-consuming and require specialized expertise, creating a bottleneck in timely diagnosis and tailored care.

This project brings together experts in psychiatry, cognitive science, machine learning, and film/media to develop a new tool that supports adult autism diagnosis through the analysis of eye movement behaviour during naturalistic viewing of social information. Our approach centres on two major aims:

Aim 1 – Stimulus Development: We will collaborate with film and media experts to produce a set of current, culturally relevant, ecologically valid videos that depict subtle, complex social exchanges. The videos will be optimized by using theories of cinematic expression and performance from film and media, and will include insights from clinical psychiatry, eye movement research, and machine learning to ensure they elicit meaningful patterns of visual attention tied to social processing.

Aim 2 – Tool Development and Validation: We will collect large-scale eye movement data from autistic and neurotypical adults while they view both our newly created, and pre-existing validated video stimuli. These data will support development of new machine learning methods for analyzing social attention by integrating multiple cues in the videos. Cues include facial expression, body language, gesture and vocal tone. This multimodal approach will enable not only the spatial identification of socially relevant regions in each frame, but also the precise temporal alignment of gaze behaviour tied to key social-emotional and auditory events during the interaction.

Our pilot data show clear differences in how autistic and neurotypical adults attend to social interactions. Despite promising pilot data, risks include difficulty integrating multimodal signals, limited effectiveness of new stimuli, and uncertain clinical acceptance. By combining expertise in psychiatry, cognitive science, film/media and machine learning, this project seeks to provide clinicians with a practical, evidence-based tool that improves recognition of autism in adults, particularly those with subtle or atypical presentations.

In cancer clinical trials, patient-reported outcomes (PROs)—which include measures of quality of life (QoL), symptoms, and functional status—sit alongside tumor response and survival as key indicators of treatment benefit. They have become increasingly important in drug development, not only for clinicians and researchers but also for regulators and health technology assessment bodies who rely on PROs when making approval and reimbursement decisions. Strong PRO data can show whether a treatment truly helps patients live better, not just longer, and can guide both benefit–risk assessments and supportive care planning. Within the broad category of PROs, QoL measures remain central, but they may not fully capture the diversity of patient experiences on their own.

In practice, most trials still rely on standardized questionnaires to collect PROs. These tools are valuable, but they have limits. They may not capture concerns that matter most to individuals—such as fatigue that disrupts family life, the financial strain of treatment, or subtle functional changes that affect daily living. Completing the same survey repeatedly can be burdensome, leading some participants to skip questions or discontinue participation. Language and literacy barriers can further limit accessibility, reducing the representativeness and utility of the data.

This project will test a different approach. We will develop a conversational artificial intelligence (AI) system, built on large language model technology, to collect PROs through a natural back-and-forth exchange—either spoken or typed. The tool will follow validated frameworks for QoL and other PRO domains, while adapting its questions to each patient’s responses, asking for clarification when needed, and exploring topics that emerge in conversation. It will generate both quantitative measures and richer narrative accounts, enabling a more complete picture of how treatment affects patients’ lives.

By combining expertise in natural language processing, psychometrics, and biostatistics, this work pushes into untested territory: conversational AI has never been deployed at scale for PRO collection in oncology. If successful, it could make PRO measurement more inclusive, less burdensome, and more informative—benefiting clinical trial design, regulatory decisions, and ultimately patient care.

Chemists design the right substance for a given task by studying the structure and properties of molecules and the changes they undergo. This is challenging because of the captivating expanse of chemical compound space and the intricacies of chemical reactions: there are innumerable potential candidates to choose from and many reaction parameters to alter. Experimental chemists navigate chemical space using their knowledge and intuition, but this trial-and-error approach is time-consuming and resource-intensive. This motivates the development of physics-based models to compute chemical properties, complementing and accelerating experimental research. However, traditional computational methods are unsuitable for screening large molecular libraries or studying complex reactions: existing models are either precise but costly or cheap and unreliable. Moreover, some properties, like cancer drug efficacy, cannot be computed at all.

This research integrates cutting-edge machine learning (ML) models and automated robotic labs to expedite the experimental and computational development of N-heterocyclic carbene (NHC) nanoclusters for cancer therapy. These stable, metal-binding compounds are attractive for their safety, tunability, and tumor-targeting potential, making them a promising platform for next-generation precision oncology. By leveraging available data on these compounds, we will train ML models to uncover patterns and capture structure-property relationships at a scale unattainable by humans. These ML models can learn from complex, high-dimensional data and systematically improve as they acquire more data, facilitated by robotic labs. Unlike current paradigms, we create a dynamic feedback loop where experimentation and computation continuously inform and refine each other, accelerating the design of next-generation cancer treatments.

Our team of mathematicians, chemists, engineers, and medical researchers will harness ML-guided predictive models and self-driving labs to design NHC-based clusters as photosensitizers for X-ray photodynamic therapy, targeting non-resectable tumors resistant to conventional chemotherapy and radiation. This is a high-risk approach because it integrates two levels of automation to tackle a practical challenge in cancer treatment, an area notorious for its high failure rates. If successful, this new paradigm will drastically cut the number of required experiments and push the boundaries of scientific exploration and discovery.

This pioneering interdisciplinary study bridges anthropology and quantum computing to explore how insecurity, in contexts of violence, displacement or climate disasters, can be more richly interpreted and modeled using quantum-inspired concepts and simulations. Its objectives are two-fold: 1) to develop quantum computing-inspired models that depict lived insecurity as probabilistic, relational and dynamic, rather than a binary state; 2) to apply these models in pilot case studies, generating insights to inform context-sensitive policy and humanitarian responses.

Drawing on Giddens’s concept of ontological insecurity (1984), the research focuses on lived experiences of insecurity shaped by structural violence, political crises, displacement and climate-related disasters, where predictability, trust and coherence collapse, often triggering flight. These insecurities are systemic and non-linear, defying the simplified models prevalent in current policy and humanitarian action.

Rooted in ethnographic inquiry, combining historical analysis, fieldwork, interviews, and IOM displacement and resettlement data, this study introduces quantum computing as a novel analytical lens to model and unravel the fluid and interdependent dynamics of insecurity. Quantum computing leverages ideas such as superposition (multiple states of insecurity coexisting, mirroring uncertain futures), entanglement (deep systemic interdependencies) and decoherence (destabilizing disruptions such as policy reversals, conflict flare-ups or disasters), offering new dynamic ways to model how insecurity emerges, spreads and transforms. Three pilot case studies will apply these ideas: 1) forced displacement and refugee resettlement as quantum state-space probabilities, 2) post-conflict trust networks as entangled systems, 3) and disaster recovery as decoherence processes disrupting ontological security. They will be modelled via quantum computing, capturing complex, uncertain social realities and generate insights for potential interventions.

This high-risk, high-reward project fuses deep, human-centred research with cutting-edge quantum computing to produce novel theory and practical simulation tools to address insecurity, enabling a nuanced visualization and anticipation of intervention outcomes in fragile settings. It opens new frontiers at the nexus of social science and quantum technology, positioning quantum computing as a groundbreaking analytical partner for anthropological inquiry.

Solar panels integrated into greenhouse coverings have long held promise as a sustainability game changer—offsetting the high energy demand and life cycle carbon intensity of year-round, climate-resilient, high-yield food production. Yet they remain rare, especially in Canada, due to technical challenges—bulky, heavy conventional panels block light and require costly load-bearing frames—and sociocultural factors affecting acceptance—e.g. worldviews and practices.

To make solar greenhouses a viable, inclusive reality for Canada’s sustainability and food security, our project takes a high-risk co-design approach to jointly optimize, for the first time, pioneering organic solar materials for crop–solar co-performance and community acceptance of solar greenhouses—a next-frontier integration hinging on reconciling competing priorities.

We will explore a new generation of thin-film, lightweight organic solar cells printable on or replacing greenhouse coverings, transmitting the light plants need and harvesting the rest for clean electricity—enabling full coverage and high power output. These organic solar cells will mark a paradigm shift for superior crop–solar co-performance—overcoming the long-standing trade-off between crop light quality and energy output and unlocking design freedoms unavailable in today’s solar technologies. In doing so, we will reimagine plant growth under solar-cell-filtered sunlight.

Yet, addressing sociocultural dimensions is critical, as technology fails without community alignment. Thus, we will continually engage farmers and Indigenous communities to co-design this technology using a Two-Eyed Seeing approach, ensuring their agricultural practices, worldviews, and environmental goals mutually inform and shape solar cell materials, devices, mounting, and crop selection. Leveraging the unmatched customizability of these solar cells into colourful, poster-like designs—unlike dark, obtrusive conventional panels—we will explore solar cell designs with farmers and Indigenous communities that align with local values and cultural heritage.

If successful, this project will forge the path for transformative solar greenhouses meeting Canada’s urgent need for sustainable food production. It will concurrently establish an anticipatory governance-based life cycle framework and a replicable, community-grounded paradigm for responsible, inclusive cleantech innovation advancing food security and energy, environmental, and cultural sustainability.

The goal of our research is to combine Indigenous history with archaeology, geomorphology, and soil science to assess how people in the Skeena River watershed (British Columbia) altered/constructed river valleys. This high-risk project challenges accepted theories and practices in numerous scientific disciplines and has two objectives: to analyze the biophysical patterns that define two salmon-bearing sub-basins; and to evaluate the role of people in shaping what are otherwise defined as natural systems, grounded in Gitxsan perspectives and epistemologies.

Gitxsan are known locally as ecosystem managers, but settler research has only recently acknowledged the role of Indigenous peoples in engineering ecosystems. As a result, Gitxsan are still characterized as hunter-gatherers, having little/no impact on their lived landscapes. These ahistorical assumptions persist to the detriment of our environment and all people in it. Indeed, cumulative decades of resource extraction, combined with colonial land-use policies, have led to the drastic watershed-level destruction of Gitxsan foods, including salmon and other cultural keystone species.

As such, innovative research is needed to examine the role of Gitxsan in actively shaping watersheds through time. We propose interdisciplinary methods to explore the archaeology, oral histories, and geomorphology of terracing and dyke features located in two Gitxsan homelands, Wii K’aax and Madii Lii. Our preliminary evidence indicates that ancestral Gitxsan physically altered major sections (20 km2) of the Sustut and Suskwa watersheds to enhance plant, animal, and fish resources; the expected significance of our work is thus high reward, as it will help to resolve pressing questions about fish, people, and environmental management.

We will use archaeological surveys, geomorphological mapping, drone-based photogrammetry, and soil geochemistry over two field seasons at both sites; we will bring together experts across 4 disciplines and combine settler and Indigenous worldviews to explicitly challenge entrenched colonial assumptions in ecology and earth sciences and develop a new theory to explain ancient Indigenous environmental management techniques. This work has the potential to enhance our understanding of historical land and water stewardship and to inform future land-use policy, salmon restoration, and cultural heritage management, enabling substantial economic, scientific, and environmental benefits for Canada.

Current diagnostic protocols rely predominantly on examination of thin tissue sections. This conventional approach necessitates sectioning of tissue samples to near-two-dimensional (2D) thickness, enabling microscopic analysis by pathologists. However, this fails to capture 3D tissue architecture, potentially overlooking critical diagnostic information. While 3D imaging technologies have been developed, their implementation remains limited due to cost and expertise. Therefore, the ability to predict clinically relevant 3D tissue features from 2D tissue sections alone and visualize these for pathologists could significantly enhance clinical workflows.

Our specific objectives are to:

1. Develop artificial intelligence algorithms to predict 3D tissue features from 2D tissue sections and evaluate model performance on several clinically relevant tasks.

2. Investigate visualization and information presentation strategies to feed this back to pathologists to understand how it affects clinical decision-making and reasoning.

We will expand on our preliminary computational model that leverages 2D histopathological data to infer 3D tumour characteristics. This exploits the stochastic nature of sectioning plane orientation, which typically bears no correlation to the spatial distribution of pathologically relevant features. By masking relevant features in the X-Y plane, we can then re-detect them in the Z (3D) plane. We will pair this with advanced cognitive science to understand how to visualize these computationally-inferred features in a way that is useful for pathologists and study how this information affects both pathologist decision making and cognitive load.

No prior studies have attempted to infer 3D tissue architecture from 2D histopathological sections nor is it known how such 3D information can be effectively fed back into the clinical setting. Consequently, this work could significantly enhance clinical diagnostic workflows, as existing technologies only profile 2D tissues and clinically relevant features in adjacent tissue planes remain undetected. This in turn could improve biomarker sensitivity and specificity, which would facilitate more accurate risk stratification and therapeutic selection. Importantly, this approach applies to existing routinely-generated datasets, eliminating the need for specialized equipment.

This project proposes a novel hybrid quantum-classical framework to enhance language models for Drug–Target Interaction (DTI) prediction, aiming to accelerate molecular discovery. The primary objective is to design and implement quantum-enhanced language models that can efficiently represent and learn the complex biochemical relationships between drug molecules and target proteins, leading to faster and more accurate identification of therapeutic compounds.

The approach integrates quantum computing, machine learning, and natural language processing into a unified system. Specifically, the project will: (1) convert molecular and protein structures into sequence-based inputs for pre-trained language models; (2) embed quantum components, such as quantum attention, variational quantum circuits, and quantum kernel estimation, within classical transformer architectures to model high-order, entangled, and nonlinear relationships; and (3) evaluate these hybrid models using benchmark and real-world DTI datasets with domain-generalization and inductive splits. A modular architecture will support benchmarking under constrained quantum hardware and enable systematic comparison to classical baselines.

This research is inherently high-risk and high-reward. It explores the frontier of applying quantum computation to biomedical AI, a space limited by hardware maturity, algorithmic scalability, and integration complexity. Yet, the potential payoff is transformative: enabling models that learn rich latent representations from biochemical data in ways not possible with classical systems alone.

The project is strongly interdisciplinary, combining quantum information theory, deep learning, cheminformatics, and biomedical data science. It brings together expertise across quantum software, molecular biology, and neural language modeling to build solutions beyond the scope of any single discipline.

The expected impact is high across three strategic areas: (1) in health, by accelerating drug development and improving therapeutic targeting; (2) in AI, by advancing models that generalize better in sparse and noisy biomedical domains; and (3) in quantum technologies, by delivering practical applications of near-term quantum computing in real-world biomedical contexts. The results could contribute to Canada’s leadership in responsible, high-impact quantum and AI innovation.

Cystic fibrosis (CF) is a life-limiting genetic disorder affecting over 4,300 Canadians, caused by a defective or absent CFTR ion channel. Despite CF being relatively well-characterized, the mechanical consequences of mucus obstruction, which can block up to 50% of the airway, remain unknown. These obstructions may exert stress on the cytoskeleton, disrupt barrier integrity, and trigger inflammation, yet no human preclinical model replicates both airway architecture and dynamic airflow to study these effects.

We propose a high-risk, high-reward interdisciplinary initiative to engineer the first 4D human artificial airway —a dynamic, patient-specific model that merges biomaterials science, stem cell biology, microfluidics, and biomechanics. This platform will enable real-time investigation of how mucus-induced mechanical stress alters epithelial function, with broad implications for CF and other lung diseases.

Aim 1: Engineering a Living, Breathing Airway Scaffold. We will fabricate optically transparent, mechanically tunable polyvinyl alcohol (PVA) hydrogel tubes with luminal biomimicking modifications and micropatterning to support long-term differentiation of airway epithelial cells under controlled airflow. Integrated into a microfluidic chip and airflow circuit, this scaffold will simulate in vivo airway dynamics, allowing visualization and manipulation of epithelial responses to mechanical stress in real time.

Aim 2: Modeling Mucostasis and Mechanical Stress in CF Airways. Using induced pluripotent stem cell-derived airway epithelium from CF patients and isogenic corrected controls, we will induce mucus production to simulate obstruction. Pressure differentials will quantify obstruction severity, while downstream assays will assess cytoskeletal remodeling, barrier disruption, inflammation, and CFTR rescue. This proof-of-concept will validate the platform’s fidelity and therapeutic relevance.

This platform reimagines how human lung diseases are studied, moving beyond static cell cultures to a dynamic, physiologically active system that mimics the structural and mechanical complexity of native tissue. The risk lies in the complexity of integrating multiple biological and physical systems into a stable, functional model. The reward is transformative as it establishes a new class of preclinical models that could revolutionize how we study CF and other pulmonary diseases.

Liquid biopsy is a non-invasive approach to sample biofluids such as blood, urine, or saliva. Within liquid biopsies, cell-free (cf)DNA – highly fragmented DNA released by both normal and diseased cells, is increasingly used as a non-invasive biomarker in diverse clinical settings, including cancer detection, prenatal testing and transplant monitoring. However, the field lacks a robust understanding of the biological and physiological factors that drive baseline cfDNA variability between individuals. Recent work from our group suggests that exercise, heart rate, and physical exertion have significant impacts on cfDNA levels, even in healthy individuals. This observation raises important concerns: how might incidental exertion — such as walking to a clinic — alter biomarker results and clinical interpretation?

This high-risk, high-reward project will address this knowledge gap through a first-of-its-kind integration of genomics, exercise physiology, and computational analysis to study the effects of physical exertion on cfDNA dynamics. Our hypothesis is that cfDNA release is highly sensitive to physiological stress, particularly mechanical and metabolic stress during exertion, with implications for biomarker timing, reliability, and interpretation.

We propose two complementary aims. First, we will use an exercise model in healthy individuals to map cfDNA release and clearance kinetics across time, intensity, and effort. Second, we will embed a contextual study in the hospital setting to assess how incidental exertion — such as walking from parking lots or climbing stairs — impacts cfDNA levels in patients of different ages and clinical conditions. Using samples from these cohorts, we will apply next-generation sequencing and computational approaches to characterize genomic cfDNA signatures associated with activity. We will apply classical machine learning and develop a multi-modal deep learning framework to identify predictive exercise-associated cfDNA signatures and assess the impact on downstream assay sensitivity and specificity.

By integrating exercise physiology, cfDNA profiling, and computational modeling, this project represents a pioneering interdisciplinary approach (high risk) with the potential to transform liquid biopsy interpretation (high reward). Leveraging sequencing, fragmentomics, and machine learning, this project will generate foundational knowledge to improve cfDNA assay performance and enhance precision in clinical decision-making.

We are a transdisciplinary team of designers, clinicians, patient partners and researchers. Our objectives are to: 1) evaluate how the Canadian health system is currently delivering back pain care, and 2) explore how care could be better delivered in the future. For feasibility, we focus on emergency departments (EDs) in British Columbia (BC).

APPROACH

Our 2-part project catalyses service design and data science methods to generate data and insights, which we will examine through ecological analysis and Akrich’s sociotechnical “scripting”—how designers anticipate interests, motives, skills, behaviours of future users and “inscribe” them into new services. Our mixed methods approach examines how back pain is constructed within and across healthcare scenarios (“scripts”). Our scripting analysis will consider ways of 're-writing' the scripts clinicians follow, to produce new scripts that better incorporate musculoskeletal specialists' skills.

In Part 1, we identify areas of greatest need for care. Using ecosystem-mapping, we will describe current back pain services and information sources. We will analyse linked health administrative data from all adults who visited an ED in BC for back pain from 2019-2023, quantifying prevalence of visits and characteristics of patients. Patients, clinicians and administrators will use the ecosystem map to trace system “touchpoints” and information transfer they have encountered in ED contexts for back pain (current-state journey-maps).

In Part 2, we co-produce new solutions for better back pain care. We will describe the sociotechnical characteristics of current ED-based musculoskeletal care models; a scoping review will identify care models. We will analyse de-identified health record data from >50,000 patients seen by a musculoskeletal specialist in the ED from 2015-2025, to characterise patients’ demographic, geographic, socioeconomic, clinical and discharge features. Using service design activities to co-produce future-state journey-maps with patients, clinicians and administrators, we will ideate and analyse solutions for what back pain care could look like.

NOVELTY & SIGNIFICANCE

The ED was designed for emergencies—not for patients who lack primary care options and have nowhere else to turn when in pain. The current "care pathway" is neither a caring nor a rational pathway. In CARE4BACKS, we apply transdisciplinary thinking and novel ways of collaborating, toward delivering a better state of back care in the future.

Background

Organ transplantation depends on rapid transfer from donor to recipient, as prolonged cold ischemia causes irreversible injury. Livers can be preserved in static cold storage for only 6–8 hours, kidneys slightly longer. Machine perfusion systems can extend this window but are bulky, complex, and expensive.

The key barrier to longer static preservation is the inability to store organs below 0 °C without ice crystal formation, which damages cell membranes. While osmotic cryoprotectants prevent crystallization in cell suspensions, they are toxic to whole organs. High-pressure cooling can also suppress ice formation, but existing systems require active pumping and are unsuitable for transport.

Preliminary Data

We have developed a passive isochoric chamber that maintains water in a liquid state at subzero temperatures through constant-volume, high-pressure conditions—without active pumping. Our prototype withstands up to 30,000 PSI (liquid water at –21 °C) and is constructed from stainless steel with an industrial high-pressure seal. In pilot studies, rat livers stored at –21 °C (30,000 PSI) for 48 h were damaged, but storage at –5 °C (~9,000 PSI) preserved viability, with function restored during subsequent normothermic perfusion.

Proposed Project

Proposed research will systematically test temperature/pressure combinations to determine the boundaries of passive isochoric subzero preservation. We will design a lighter chamber operating at lower pressures, with integrated sensors and an embedded camera for real-time observation, and capacity for multiple rodent organs to increase throughput. The chamber will be built using a combination of additive manufacturing and CNC machining.

Impact

This work will establish the feasibility of a simple, low-cost, portable method to store transplantable organs for multiple days—overcoming current time and distance barriers in organ allocation and enabling long-range transport for life-saving transplants.

Many modern, high-value pharmaceuticals originate from Indigenous Knowledge (IK), yet Indigenous Peoples are often excluded from development processes and benefits, particularly in genomics and biotechnology. Canada's lack of regulations leaves communities vulnerable to biopiracy, while researchers lack ethical guidance. This interdisciplinary project examines barriers and opportunities for meaningful Indigenous engagement in pharmaceutical and nutraceutical development derived from IK. We will review and develop frameworks for Indigenous participation across the value chain—from IK recognition to biotechnological production and commercialization. As a case study, we will use montbretin-A (MbA), a natural product in Phase-1 clinical trials for type 2 diabetes originating from a plant native to Southern Africa. This will be Canada's first comprehensive study of Indigenous engagement in pharmaceutical development, combining Indigenous methodologies with genomics and biotechnology. Our participatory approach addresses critical ethics gaps, while fostering equitable, reciprocal relationships with Indigenous communities in Canada and South Africa.

Over two years, we will: (1) review international best practices in Access and Benefit Sharing, including Nagoya Protocol implementation, and current Canadian legal frameworks; (2) conduct an environmental scan of community-based enterprises in Canada focused on IK-derived therapeutics; (3) build relationships with Indigenous communities, including cross-cultural knowledge sharing with traditional medicine practitioners to understand historical experiences and engagement preferences; (4) hold key informant interviews with community leaders, researchers, knowledge holders, and industry; (5) analyze the MbA case to identify traditional plant use and engagement opportunities in cultivation, biotechnology, clinical trials, and commercialization; (6) develop engagement frameworks, guidelines, and toolkits for researchers and industry; (7) establish relationships with international scholars.

This project tackles biopiracy concerns while creating pathways for Indigenous benefits from IK-derived medicines through knowledge reciprocity protocols. Outcomes include engagement frameworks and guidelines, stronger global relationships with academics and traditional knowledge holders, enhanced transdisciplinary research ties, and models for ethical cross-cultural knowledge exchange to inform future national regulatory frameworks.

Batteries are the pinnacle of a zero-emissions future, enabling electrified transportation and large-scale storage of intermittent renewable energy sources. As demand for ultrafast charging grows—with the proliferation of electric vehicles, portable electronics, and grid-scale energy infrastructure—a persistent bottleneck remains: the sluggish transport of lithium ions (Li⁺) across the solid electrolyte interphase (SEI). This nanometric interface, formed between the anode and electrolyte, often comprises low-conductivity inorganic compounds such as LiF, Li₂CO₃, and Li₂O, which impose high activation barriers (0.3–0.7 eV) to Li⁺ migration. Despite decades of research on tuning SEI composition and morphology, a fundamental shift in how we approach interfacial ion transport is needed to achieve step-change improvements in charging speed.

This exploratory proposal introduces the concept of Quantum Tunneling Interfaces (QTIs): engineered layers that leverage quantum mechanical tunneling under electric field modulation to enhance Li⁺ transport across interfacial barriers. Unlike conventional SEI engineering strategies that focus on lowering diffusion barriers through bulk phase optimization, QTIs are envisioned as nanoscale structures with tailored roughness, electrostatic potential gradients, and microstructural heterogeneity designed to activate and direct Li⁺ tunneling. These interfaces may serve as dynamic ion injection reservoirs, accelerating the transition of Li⁺ from the electrolyte into the anode, potentially bypassing rate-limiting desolvation during battery charging.

The research will combine computational and experimental approaches to validate the QTI hypothesis. Quantum transport modeling (e.g., Non-equilibrium Green's function density functional theory (NEGF-DFT)) will be used to simulate tunneling conditions under applied fields. Ab initio molecular dynamics (AIMD) and DFT calculations will assess interfacial stability and field effects. Experimentally, QTI prototypes will be fabricated and characterized using in situ spectroscopy and electrochemical techniques to measure ionic conductivity and charging dynamics.

If successful, this project could fundamentally redefine the role of the SEI, enabling batteries that charge in minutes rather than hours. It also opens a new frontier at the intersection of quantum mechanics, electrochemistry, and materials science—representing the kind of high-risk, high-reward research envisioned by this program.

Canada lacks a Made‑in‑Canada lifecourse policy simulator that links nutrition as the exposure in childhood to adult chronic disease, health equity, and public‑payer healthcare costs. LIFE‑Canada (LIFEcourse Impact & Equity Evaluation Engine) will provide local, decision‑grade estimates calibrated to Canadian demography.

Objectives: (1) design a child‑first lifecourse architecture and population engine; (2) build and validate paediatric‑to‑adult disease and cost kernels; (3) quantify structural uncertainty and equity impacts for priority nutrition policies; and (4) deliver a policy‑mix optimiser and public dashboard for rapid scenario testing.

Approach: This inherently interdisciplinary project integrates public/population health, nutritional epidemiology, health policy and economics, biostatistics and demography, and computer and information sciences (reinforcement‑learning). We will develop a modular, open‑source microsimulation with a demographic foundation aligned to national projections and a child state vector that moves beyond body mass index (including diet quality, adiposity, physical activity). Calibration and validation will use Canadian population surveys and existing health–administrative linkages. Policy analyses will employ constrained reinforcement learning (RL) with a reward that combines net health benefit (health‑adjusted life‑years), public‑payer healthcare costs, and an equity‑shortfall penalty. Where point identification is unjustified, unit‑selection bounds will rank jurisdictions and equity strata transparently. All code and synthetic data will be open.

Novelty and significance: LIFE‑Canada will be the first Made‑in‑Canada, child‑first lifecourse nutrition‑policy simulator with explicit equity constraints and demographic anchoring. By producing credible, local estimates of health gains, healthcare costs, and distributional effects, it enables federal and provincial decision‑makers to compare alternative policy bundles (e.g., child‑directed marketing restrictions, front‑of‑package symbols, and sugary‑drink taxes) and choose options that maximise health‑adjusted life‑years, reduce public‑payer costs, and narrow equity gaps. The platform generalises to other risk factors and establishes a durable national capability for transparent, evidence‑based policy evaluation.

Obstructive sleep apnea (OSA) is one of the most common yet underdiagnosed sleep disorders, affecting millions of Canadians. Left untreated, OSA increases the risk of heart disease, diabetes, and reduced quality of life. Current diagnostic methods rely on resource-intensive sleep studies, which are inaccessible for many, especially those in rural and underserved communities. At the same time, genetic factors play a major role in OSA, but are rarely incorporated into clinical care. Our project tackles this gap by integrating genomic risk information with wearable sensor data, enabling new ways to detect, monitor, and manage OSA outside of hospital settings.

The objectives of this project are threefold: (1) to develop and validate a genetic risk panel for OSA using recent discoveries from large-scale genomic studies; (2) to combine this panel with real-world wearable data to create an integrated tool for early risk stratification; and (3) to design an ethical, artificial intelligence–driven framework that can guide diagnosis and follow-up care while safeguarding privacy and equity.

Our approach is deliberately high risk. Combining genomics, wearable technologies, and AI poses major scientific and ethical challenges, including the integration of complex datasets, ensuring interoperability across platforms, and addressing concerns about consent, bias, and re-identification. However, the potential rewards are transformative. If successful, this work will establish a new model for precision sleep health that is scalable, cost-effective, and accessible. By embedding ethics directly into system design, we aim to ensure that the resulting innovations benefit diverse populations equitably.

This two-year multidisciplinary collaboration between medicine, data science, and information systems will generate tools and frameworks that can inform future large-scale clinical studies and health policies. Ultimately, our project seeks to move beyond the walls of the sleep laboratory, offering a pathway to earlier detection, personalized care, and healthier lives for people living with OSA.

There is an urgent need to both reduce CO₂ production and capture existing emissions. The world releases 35 to 40 gigatonnes (Gt) of anthropogenic CO₂ into the atmosphere every year, yet globally, we are currently capturing only about 0.1% of that, which is a mere fraction compared to the 10 Gt per year of removal necessary to keep global warming below 1.5°C by 2050. To return to a relatively “safe” atmospheric CO₂ level of around 350 ppm, we need to remove approximately one trillion tonnes of legacy CO₂, a goal that remains far out of reach for society. Additionally, effective carbon management is crucial for addressing emissions from some of the toughest-to-abate sectors, such as heavy industries like steel and cement production. It’s clear that, in parallel with transitioning away from fossil fuels to renewables, we must also advance carbon capture technologies.

However, this development is not just a technological or engineering challenge; it demands an integrated approach that considers social, political, sustainability, and economic dimensions. For instance, from a sustainability perspective, what happens if the carbon capture technology uses materials whose production results in significant emissions? From a social perspectives, how can governance models be designed to prevent this technology from being used as a “greenwashing” tool by industries that delay meaningful emissions reductions while only capturing a small fraction of their CO₂ output? Economically, if the process isn’t viable, it may never be widely adopted by industry, regardless of its actual effectiveness. These complexities highlight the necessity of a truly holistic approach.

This project ambitiously aims to tackle these intertwined challenges by combining engineering, social, sustainability, and economic expertise focused on developing a promising carbon capture technology based on an electrochemical, organic-based system that operates at room temperature and ambient pressure and can be integrated with renewable energy sources. The goal is to create a technology that not only meets technical challenges but is also sustainable (biomass-derived), economically viable, and positioned in a way to avoid becoming a justification for continued reliance on fossil fuels. While this is a high-risk and ambitious endeavor, the potential reward is significant: tackling the greatest challenge of our time by reducing anthropogenic CO2 emissions and mitigating climate change.

The proposed research directly addresses the aspirations for culturally grounded, community-led approaches to improving the

wholistic wellness of Indigenous youth. By centring traditional carving practices and intergenerational teachings, this project

operates at the intersection of cultural resurgence, youth empowerment, and wholistic health and wellness promotion. The

program embraces a strengths-based, transdisciplinary framework that integrates Indigenous knowledge systems and

community leadership with evidence-informed methodologies. This aligns with national calls for culturally safe, inclusive, and

sustainable health and wellness initiatives that are shaped by and for Indigenous communities.

Through experiential learning guided by Elders and Knowledge Keepers, the research will explore how reconnecting with

traditional practices contributes to youth mental, emotional, physical, and spiritual well-being. It also addresses knowledge

gaps in how culturally embedded activities foster resilience, identity formation, and intergenerational healing. Furthermore, this

research responds to ongoing concerns around systemic barriers in healthcare and education by creating a culturally relevant

space that uplifts Indigenous voices and ways of knowing, being, and doing. The findings will inform policy, practice, and

program development aimed at supporting Indigenous youth through culturally meaningful, land- and art-based wellness

initiatives.

This interdisciplinary research project addresses an urgent and underexplored question: how might emerging deep-sea mining (DSM) operations jeopardize the profitability, market stability, and long-term viability of Pacific fisheries by introducing contamination risks to high-value seafood products? As industrial interest in DSM grows, particularly in the Clarion-Clipperton Zone (CCZ), a 4.5-million-km² region with over 18 exploration contracts, concerns have mounted over the dispersal of toxic metals such as cadmium, cobalt, and nickel. These contaminants could bioaccumulate in marine food webs, undermining seafood safety and fisheries sustainability. This project pursues three integrated objectives: (1) to estimate seafood contamination risks by modeling metal exposure pathways and bioaccumulation in key commercial species (e.g., tuna, shrimp, and benthopelagic fish) under varying DSM intensity scenarios; (2) to evaluate economic losses to fisheries, including revenue and profit reductions, public health costs, and the erosion of consumer confidence in seafood safety; and (3) to develop spatial seafood risk maps that integrate biological contamination modeling with economic vulnerability indicators, supporting evidence-based decision-making for ocean resource governance.

Our research approach combines ecosystem modeling (Ecopath with Ecosim, Ecotracer module), environmental toxicology, and economic loss simulation. Contaminant input data will be sourced from DSM contractor environmental assessments (e.g., NORI-D), and seafood tissue concentrations will be estimated using published bioaccumulation factors. These will be compared to international food safety thresholds (e.g., Codex). Economic impact will be assessed by applying modeled contamination risk to regional fish catch data and public health burden estimates, incorporating a sensitivity analysis to simulate 50%, 80%, and 100% contamination scenarios. GIS tools with Spatiotemporal Bayesian analysis and probabilistic Multi-Criteria Decision Analysis (MCDA) will be used to create spatial risk maps identifying ecological and economic vulnerability hotspots across the CCZ.

This high-risk project uses novel modeling in uncharted deep-sea ecosystems, combining toxicology, economics, and spatial analysis to deliver the first integrated tool for assessing fisheries impacts from deep-sea mining—supporting global ocean governance.

Canada’s agri-food activities are essential parts of its economy, contributing approximately 7% of Canada's GDP and $149.2 billion in activities in 2024. This sector faces multiple, overlapping risks due to climate change, market instability and extreme weather events. Increased fire activity in recent years is reshaping the future of the Okanagan Valley's wine terroir—comprised of natural regional characteristics such as climate, soil, and topography moulded by human craftsmanship. Geographically indicated terroir is equated with quality in wine-producing regions which is crucial to commercial success in the highly competitive market. Okanagan appellations contain 86% of BC’s vineyard acreage, making it the dominant part of BC’s $3.75 billion annual wine industry.

We propose to study how the Okanagan Valley terroir is perceived and understood by the market in the context of a now persistent problem in the wine industry: smoke-taint. An interdisciplinary approach will intertwine consideration for the terroir with conceptually novel ways of photochemically manipulating the aroma/flavor of wine. In the Okanagan Valley, the marketing of terroir is rooted in the natural surrounding plants. The Okanagan is a nascent new world wine region, distinguished from old world wines that have been prepared traditionally for generations. A risk trade off introduces the primary research question: is it is possible for a new world wine region built on biodiverse nature to compete in an international market imbedded in tradition, by using high-tech processing to sustain or even redefine terroir? Would high-tech processes save the Okanagan wine industry, or mask its true terroir?

Bio-based interventions have gained acceptance in the wine industry, yet questions remain about the acceptance of high-tech chemical processing. Smoke-taint in wine is caused by molecules called volatile phenols (VPs), pungent compounds produced by wildfires. Affected grapes don’t have unpleasant smoke-taint aromas until they are fermented and these VPs are released, lowering the value of the produced wine. This project will explore the use of carbon nitride, an organic semiconductor, as a solid material photocatalyst that harnesses sunlight to chemically degrade VPs and remove the unwanted flavors. In developing novel avenues for manipulating wine chemistry and hence aroma, we will explore the concept of terroir with respect to quality of the air and high-tech interventions in wine-making.

Continuous capture and conversion of waste CO₂ from large-scale chemical processes into value-added chemicals and fuels presents significant opportunities for process optimization, chemical circularity, and environmental remediation. Reactive carbon capture (RCC) is an emerging technology that integrates CO₂ capture and conversion into high-value chemicals (e.g. methanol, CO, and C1+ hydrocarbons), offering scalability, cost-effectiveness, and the potential for net-zero CO₂-to-fuel cycles. However, current RCC technologies face critical barriers related to material costs, poor durability of sorbents and catalysts, and the energy-intensive conditions required for operation.

This interdisciplinary project proposes a high-risk, high-reward approach to overcome these limitations by developing the first light- and microwave-activated hybrid porous organic nanomaterials (HPONs) for RCC. By uniting expertise from materials science, photochemistry, microwave engineering, renewable energy, and artificial intelligence (AI), this project will pioneer an entirely new RCC platform powered by renewable energy and non-thermal mechanisms. HPONs will encapsulate CO₂-selective polymers and advanced plasmonic photocatalysts within transparent, porous beads. The polymers will capture and concentrate CO₂, releasing it in response to targeted microwave energy, while the embedded plasmonic nanoparticle catalysts will convert CO₂ into fuels under UV-visible light through non-thermal charge carrier generation.

This unconventional integration of light-and microwave-activated materials for CO₂ capture and conversion has never been explored and represents a clear departure from conventional RCC methods. The project's high-risk elements lie in its novel, untested materials platform, the challenge of coupling multiple non-thermal energy inputs, and the ambitious goal of integrated, energy-efficient CO₂ conversion.

The potential high reward includes a paradigm-shifting technology capable of significantly reducing the energy and material costs of RCC, enabling scalable deployment, and accelerating Canada's decarbonization efforts. Crucially, the use of AI/ML for accelerated materials discovery and system optimization introduces an unprecedented opportunity to radically enhance the speed, efficiency, and effectiveness of RCC innovation; further amplifying the project's transformative impact. This work advances materials chemistry and renewable energy integration in Canadian economy.

Plant diseases cause over US$220 billion in crop losses annually, posing a significant and growing threat to Canadian agriculture and food security. Current diagnostic approaches focus on identifying individual pathogens, but this fails to address the unpredictable variation in disease phenotypes. For virus diseases, the same virus can cause severe or mild symptoms for reasons that are poorly understood. We propose a paradigm shift, moving beyond cataloguing viruses, towards understanding complex social interactions within the entire community of viruses in a host or ecosystem, known as the virome.

Our project introduces a novel, interdisciplinary framework that merges applied field work (plant pathology) with theory (social evolution) to test a high-risk, high-reward hypothesis: that the severity of crop disease is driven by the balance of cooperation and cheating within the virome. Viruses are social entities. Many viral infections consist of a mix of complete 'cooperative' genomes and defective 'cheater' genomes that rely on cooperators to spread. We predict that the relative abundance of these cheaters is a key determinant of disease development and phenotype.

Our objectives are:

1) to develop a sequencing workflow to efficiently uncover viromes of Rosaceae and Musaceae plants;

2) to develop bioinformatics pipelines to quantify the composition of cooperator and cheater sequences in our virome datasets; and

3) to develop mathematical models to examine how virome structure and interactions between cooperators and cheaters could correlate with disease phenotypes.

We will focus on Rosaceae species such as apples (Malus spp.) and pears (Pyrus spp.), which contribute almost 2,500 jobs and over CA$160 million to British Columbia’s economy. We will leverage our extensive longitudinal collection (>1000) of diseased Philippine Musaceae species (banana (Musa spp.), abaca (Musa textilis)), which are the country's top exports, contributing over US$1.3 billion to the country's economy.

If successful, this research would represent the first empirical testing of social evolution theory in a natural tropical and temperate plant-virus system. It would provide a new way to predict disease risks. It may result in groundbreaking solutions, such as using 'cheater' virus genomes or particles as a form of 'virome therapy' to manipulate disease outcomes. This method would help protect crops from disease outbreaks and maintain agricultural productivity and food security.

Canada is entering a critical phase in regulating per- and polyfluoroalkyl substances (PFAS), a class of persistent toxic compounds. While only PFOS, PFOA, and select long-chain perfluorocarboxylic acids are listed under CEPA Schedule 1, a Mar. 2025 federal report concluded the broader PFAS class (excluding fluoropolymers) meets CEPA toxicity criteria. A class-based risk tool is in development and may set PFAS limits for wastewater and biosolids. Interim measures include a 50 µg/kg PFOS biosolids limit and Québec’s 2025 code with tighter thresholds. The Wastewater Systems Effluent Regulations (SOR/2012-139) still lack PFAS limits, reflecting the absence of scalable treatment options. PFAS enter wastewater from industry, consumer products, and landfills, persist through secondary treatment, and partition to effluent, aerosols, and biosolids. About 58% of Canadian biosolids are land-applied, risking soil and groundwater. A Feb. 2025 case in Texas linked PFAS-contaminated biosolids to livestock deaths and surface water pollution, underscoring the need for upstream destruction.

We propose a modular photonic sleeve: Electrospun PVC nanofiber cartridges that couple PFAS capture with sequential degradation. PVC is selected for durability and functionalization, producing high-surface-area fibers with tailored chemistries that stabilize bubble films and sorb PFAS via hydrogen bonding, halogen bonding, and electrostatics. This enables foam-phase preconcentration at air-water interfaces where PFAS are enriched. Side-emitting optical fibers activate a photocatalyst for cleavage of C-F bonds, while an outer biofunctional layer with immobilized dehalogenases and oxidoreductases provides enzymatic finishing from biotransformation to biodefluorination. The modular platform fits within aeration basins and sidestream channels before clarification, intercepting PFAS where concentrated and integrating with existing hydraulics.

This high-risk, high-reward platform integrates three rarely combined mechanisms to enable true in situ PFAS destruction. It addresses regulatory and technical gaps by moving beyond removal to destruction, eliminating secondary wastes, and providing measurable defluorination endpoints aligned with emerging standards. By converging materials science, photocatalysis, microbiology, enzymology, biogeochemistry, and process engineering, the platform links molecular mechanisms with plant-scale deployment for a transformative PFAS treatment strategy.

Deep space exploration is a novel frontier. The continuity and success of missions will depend on research into women’s space health and reproductive health. Initial in vitro studies of mouse embryos and in vivo studies of mouse pregnancies demonstrate altered embryonic growth and reduced placental development, pointing towards potential adverse outcomes in future human space pregnancies. Characterizing the impact of spaceflight and microgravity on a placenta-on-a-chip (PoC) system will set the stage for understanding maternal health in space and inform the need for future reproductive technologies or therapeutics.

We propose novel technologies to evaluate the impact of spaceflight and microgravity on human pregnancy. With no current studies of human pregnancy in microgravity, and ethical and logistical barriers to such research, the development of PoC technology represents an opportunity to assess the viability of long-term deep space missions. A multidisciplinary team consisting of a biomedical engineer and nanotechnologist with expertise in Organ-on-a-Chip (OoC) design, a placental biologist with expertise in placental pathologies, and an aerospace researcher will drive the development of an autonomous PoC platform for spaceflight. To our knowledge, this will be the first Canadian-led OoC system for spaceflight, and the first worldwide to investigate in-flight human pregnancy.

The project will progress in three phases: Phase I will evaluate the effects of simulated microgravity on the placenta; Phase II will assess the technological readiness level of the autonomous PoC platform through parabolic flight testing; and Phase III will conduct parallel PoC experiments in Earth gravity, simulated microgravity, and in-space microgravity using low-orbit satellites. Beyond significantly advancing knowledge of maternal space health, this platform will help establish Canadian leadership in medical space research by enabling further studies on diverse tissues and organ models in microgravity. An automated high throughput OoC system will also reduce the costs of in-flight experimentation. On the Canadian front, this can also help the development of reproductive technologies for future moon and Mars missions, which can translate to reproductive technologies for remote Canadian communities. The development of self-sustaining medical technology will be of benefit to both future space missions and remote Canadian communities.

Gene expression in metazoans is exquisitely controlled in space-time by cis-regulatory elements (CREs), accessible regions of the genome cooperatively bound by transcription factors (TFs).

Naturally, regulatory precision mediated by the binding of TF to chromatin cannot be infinite: biophysico-chemical constraints limit the sensitivity and specificity of CREs’ response to small perturbations in cellular states. For instance, increase of a single TF molecule per cell presumably cannot be resolved by the regulatory machinery because of thermal fluctuations battering all components. While delineating limits of chemo-sensing has a long history in model organisms (e.g., E. coli chemotaxis, Drosophila gap gene patterning), its exploration in higher vertebrate’s gene regulation remains sparse. This is in part because of the combinatorial complexity of the cis-regulatory code, which obscures the quantitative nature and stringency of these bounds. Since biology only samples a tiny part of possible cell-states, non-perturbational measurements cannot resolve this fundamental question.

Combining biophysics, chromatin genetics, and functional genomics, we will measure CREs’ response to precise modulation of TF expression in diverse mammalian cell types. First, graded TF expression changes will be induced using CRISPR-activation after differentiating stem cells to multiple target cell types. Concurrent assessment of chromatin accessibility using high-content single-cell co-assays will identify CREs that are TF-inducible in a lineage-specific manner. Second, to comprehensively map regulatory trade-offs, we will use biophysical models of TF-chromatin interactions to design CREs optimizing two conflicting objectives: (1) sensitivity: maximizing CRE’s response to individual TF induction, and (2) specificity, i.e., strong dependence on the cell-state. Performance of hundreds of thousands of optimized CREs will be quantified using multiplex reporters. Our hypothesis is that key biochemical sequence features will map to these functional phenotypes (e.g., sensitivity – number/affinity of TF binding sites; specificity – combinations of induced and lineage-restricted TFs).

This constrained reverse-engineering will identify actionable rules and bounds on CRE customization, with transformative implications for gene therapy design as well synthetic biology applications requiring fine expression control across multiple mammalian cell types.

From a single cell, an entire animal body takes shape: proportionate, organized, and robust. How this astonishing transformation unfolds remains one of biology’s most enduring and inspiring questions. A particularly striking feature is scaling: the ability of embryos of different sizes to produce the same body plan, precisely rescaled. This phenomenon spans nature, from flies and frogs to stem cell–derived synthetic systems. Yet, despite its ubiquity, the mechanisms that enable biological systems to scale remain largely unknown.

Recent work from the Gregor lab revealed a surprising feature of gene expression dynamics during early fly development: gap gene patterns evolve along a low-dimensional trajectory, suggesting latent regulatory modes. One such component, which we call “mode 0”, may serve as the foundation of scaling. Its origin, function, and predictive value remain unexplored. Understanding mode 0 could uncover new principles of pattern formation and dynamic control in development.

This high-risk, high-reward project will investigate how mode 0 arises and whether it enables proportional patterning. On the theoretical side, the François group will develop generative models of gene regulatory dynamics constrained by scaling requirements, using differential programming, machine learning, and dynamical systems theory. This will enable the construction of synthetic developmental systems and test whether scaling emerges generically or requires a fine-tuned architecture. These models will be tested against high-resolution empirical data: the Gregor lab will perform systematic perturbations of maternal and zygotic inputs in fly embryos and quantify their effects on gap gene dynamics and mode structure. We will analyze how speed, robustness, and identity of dynamical modes depend on upstream regulators, focusing on whether mode 0 arises from maternal gradients, network structure, or feedback.

This project is inherently interdisciplinary, merging quantitative developmental genetics, live imaging, statistical physics, and differentiable simulation. While physics and biology have intersected, our approach uses tools rarely applied in development to ask fundamentally new questions about biological organization. This work could redefine how we understand morphogenetic scaling, reveal new design principles of robust pattern formation despite molecular variability, and inform the engineering of scalable synthetic tissues or embryos.

Context -- Recently, the G7 summit in Kananaskis recognized that quantum computing was likely to "bring significant and transformative benefits to societies worldwide." However, to build a useful universal quantum computer, it is widely believed that quantum error correction will be required to protect devices from decoherence. Bosonic codes, where qubits are encoded in harmonic oscillators such as microwave cavities, have proved a promising approach. For example, experimental demonstrations using simple instances of the Gottesman-Kitaev-Preskill (GKP) family of codes have achieved performance beyond the break-even point, that is, where error correction yields more reliable quantum information than leaving it uncorrected.

At the same time, in a completely different field, mathematicians have spent over two centuries studying geometric objects called abelian varieties. These shapes have rich internal structures and have been central to many major advances in pure mathematics, from the theory of elliptic functions to modern geometry and number theory.

We recently made a surprising discovery: the exact same information needed to define a GKP code is also what defines an abelian variety. This unexpected connection opens the door to new and potentially powerful interactions between mathematics and quantum computing.

Project goal -- We aim to build a bridge between these two seemingly distant worlds: the deep mathematical theory of abelian varieties and the practical challenges of quantum error correction. Our goal is to leverage two centuries of research on abelian varieties to guide the design and analysis of GKP codes. To this end, we will investigate questions such as: Which codes offer the greatest robustness against errors? Which enable the most efficient implementation of logical operations?

Expected significance -- This project is timely, with academic groups (MIT, ETH Zurich) and industries (Nord Quantique) currently pursuing GKP codes as a promising architecture for quantum computing. Our approach could inform code design while uncovering new questions in the theory of abelian varieties. It may also apply to quantum sensing, where GKP codes are increasingly studied.

Although mathematics and physics have a long history of interaction, the theory of abelian varieties has remained far removed from quantum computing. This project brings the two fields together in a way that requires, and benefits from, close collaboration across disciplines.

The human and financial burden of major psychiatric diseases (MPDs) with a neurodevelopmental component—including schizophrenia, bipolar disorder, and recurrent major depressive disorder—affecting over 10% of the population is considerable. Although these disorders clearly involve neurodevelopmental mechanisms, their etiology remains poorly understood, and available treatments are essentially palliative.

Emerging evidence suggests that subtle disturbances of the vestibular system (VS) are present in patients with MPDs and may provide valuable insights for both diagnosis and risk screening in their offspring, through a better understanding of early pathogenesis. The VS plays a cardinal role in multisensory integration—an essential process that is consistently impaired in MPDs. These disturbances may thus represent a developmental endophenotype highly predictive of MPD onset. Promising preliminary findings using a novel electrophysiological technique, Electrovestibulography (EVestG), have already revealed alterations of VS function in some MPDs.

Based on these findings, we propose to investigate VS functioning in patients with MPDs and in their high-risk (HR) offspring, in whom such alterations could contribute to multisensory and bodily self-disturbances. We will recruit HR children (6–24 years) with a parent affected by an MPD, as well as adult patients with MPDs. A cross-sectional study will combine EVestG recordings, VS functional assessments, multisensory integration tasks, and measures of bodily self-disturbance.

Our multidisciplinary team brings together complementary expertise in longitudinal psychiatric studies (Marquet), vestibular system research (Sharp), and EVestG signal analysis (Lithgow, Moussavi). The study objectives are:

To characterize VS disturbances in relation to multisensory integration and bodily self-awareness, as putative biomarkers of MPDs, particularly in HR children;

To develop novel bio-signal analysis algorithms tailored for risk screening in youth, including methods to disentangle the effects of comorbid anxiety/depression and medication influences, using interpretable AI approaches.

Because deficits in multisensory integration are a core feature of MPDs, this research project will contribute to elucidate key mechanisms of their pathophysiology and developmental trajectory, paving the way for innovative therapeutic strategies with an emphasis on prevention in children at risk.

Les monomères cycliques d'acides gras (MCAG), présents en traces dans les huiles et graisses alimentaires

transformées (ex. friture, raffinage, hydrogénation), font partie de l'alimentation humaine. Des études chez

l'animal indiquent qu'à ces très faibles concentrations, les MCAG sont pro-athérogènes, causent une

accumulation de triglycérides dans le foie, altèrent le métabolisme des enzymes hépatiques et induisent une

stéatose. Ces effets des MCAG apparaissent à des doses nettement inférieures à celles des acides gras trans,

interdits par Santé Canada. Cependant, les études sur les gras trans ne rapportent pas les teneurs en MCAG

présents dans ces graisses, même s’ils ont vraisemblablement contribué aux effets santé attribués aux gras

trans. Cette absence des teneurs en MCAG dans les études peut être due à divers facteurs tels que leur très

basse teneur ou même l'ignorance de leur présence dans les huiles transformées. En effet, les MCAG

demeurent méconnus et, à des teneurs pro-athérogènes, passeront inaperçus dans une analyse de routine. Par

ailleurs, les données de la littérature scientifique ne sont pas concluantes sur un lien dose-réponse pour les

effets observés sur la santé avec les acides gras trans industriels. En combinant ces informations suggérant que

les MCAG contribuent aux effets négatifs sur la santé attribués aux acides gras trans, nous émettons les

hypothèses suivantes : 1) la littérature scientifique et les bases de données publiques contiennent des

informations permettant d’estimer la teneur en MCAG dans les gras trans utilisés dans les études ; 2) à partir

de ces données, il est possible de construire une démarche analytique permettant d’estimer la part des MCAG

dans les effets santé observés et d'expliquer l’absence d’un lien dose-réponse des gras trans ; 3) le retrait de

l'alimentation des gras trans et leur substitution par des huiles insaturées aura un effet mitigé sur la santé des

canadiens.

Avec ce projet à haut risque, nous remettons en question le paradigme selon lequel les acides gras trans seuls

sont nocifs pour la santé et proposons que les MCAG, présents à l’état de trace dans les aliments transformés,

sont également nocifs. Ce projet de recherche interdisciplinaire offre un potentiel de rendement très élevé, car

une meilleure compréhension des impacts des graisses transformées sur le métabolisme des lipides entraînera

des changements dans les politiques nutritionnelles, assurant ainsi une meilleure santé à la population.

Light is a universal scientific tool for probing, characterizing, and modulating physical and biological systems. Because photons can interact non-invasively and be detected with exquisite temporal, spatial, and spectral precision, light has become indispensable in fields as diverse as biophotonics, astrophysics, and remote sensing. In astrophysics and molecular imaging, light is used not just to illuminate systems, but to extract information from weak or noisy signals that would otherwise remain hidden. For example, while gravitational wave (GW) observatories rely on laser interferometry to detect minute distortions in spacetime caused by distant cosmic events, super-resolution microscopy (SRM) uses advanced photon-based techniques to resolve nanoscale molecular events in biological systems. These domains share the same fundamental challenge of detecting extremely faint signals embedded in complex noise environments, wherein only a small number of photons or minute variations in their behavior carry critical information.

This project proposes to co-develop a new transdisciplinary framework for sub-threshold signal detection. By integrating time-frequency analysis and noise mitigation methods from GW science with probabilistic models of photon emission in SRM, we aim to transform the ability of both fields to detect events currently hidden beneath noise floors.

We will apply these advances to gain new insight into synaptic alterations in depression for which the underlying molecular mechanisms remain poorly understood. At the heart of synaptic signaling are protein complexes and receptor nanodomains whose assembly, disassembly, and mobility are critical to maintaining adaptive neurotransmission and can be revealed by SRM. In GW astrophysics, insights from microscopy, such as optimizing detection with very few photons could inform the next generation of weak-signal search pipelines, particularly for signals with incomplete models such as core-collapse supernovae or eccentric mergers. We will also apply novel statistical methods to enhance signal and noise classification, with an emphasis on scientific explainability.

This project seeks to define a generalizable science of weak signals, catalyzed by the convergence of challenges in astrophysics and biophotonics. Beyond the field-specific technological advances, it will ignite a conceptual shift in how researchers across widely varying domains approach the fundamental challenge of detecting faint signals.

Context: In Canada, approximately 15,000 infants are affected annually by prenatal substance exposure. Pregnancy substance use is linked to numerous adverse outcomes, including congenital disabilities, neurodevelopmental delays, and long-term learning difficulties. The detrimental impact of prenatal drug exposure worsens with polysubstance use, particularly involving opioids. Thus, understanding the molecular and cellular effects of polydrug exposure is crucial.

Over the last decade, organoids, which are self-organizing three-dimensional (3D) tissue cultures derived from stem cells, have emerged as promising models that mimic key structural, functional, and biological features of organs. Since they preserve the genetic background of the donor, they can be used to track the trajectories of neurodevelopmental development of neuronal networks.

Brain-specific organoids are generated from induced pluripotent stem cells (iPSCs) or embryonic stem cells (ESCs) and directed toward neural lineages with developmental cues to represent better the brain's complexity, cellular diversity, and neural architecture.

Exposure to drugs like ethanol, morphine, and nicotine has been shown to alter neuronal differentiation, increase cell death, and cause metabolic issues. However, the effects of combining multiple drugs remain untested.

Objectives and hypothesis: This project aims to investigate the molecular and cellular changes caused by early drug exposure. We hypothesize that polydrug exposure will produce specific effects related to drug combinations. We will study how the neurodevelopmental trajectories of human iPSC-derived neurons are affected by polysubstance exposure.

Our objectives include: 1. Conducting dose-response experiments with human iPSC-derived neurons, exposing them to ethanol, nicotine, morphine, and their combinations to assess impacts on neuronal activity and network formation. 2. Generating assembloids from midbrain and striatal organoids following drug of abuse exposure. Functional imaging will analyze neuronal activity, axonal projections, and neurotransmitter functions. 3. Applying genomic approaches to identify changes in the epigenome to compare results with publicly available datasets.

Impact: Our innovative model will provide a human-relevant platform to clarify how polydrug exposure affects neuronal development, paving the way for future research and the development of novel therapies to address the long-term effects of such exposure.

Wildfire seasons in Canada are becoming longer, warmer, and more destructive due to climate change. In recent years, these fires have released more greenhouse gases than the entire transportation sector of the country in a single year. These events threaten lives, ecosystems, infrastructure, and economies, while straining emergency services and healthcare systems. This project aims to create a fully integrated decision platform that manages wildfires throughout their entire lifecycle, encompassing prevention, detection, suppression, evacuation, and recovery, to protect people, reduce emissions, and enhance resilience. Our approach combines predictive wildfire modeling, real-time logistics planning, and population health analysis into a single forecast-to-action pipeline. Machine learning models will anticipate ignition risk, fire growth, and smoke dispersion. Operations research methods will optimize the placement and movement of firefighting crews, aircraft, equipment, and evacuation routes under uncertainty and resource constraints. Recovery planning will sequence safe re-entry, restoration activities, and public health interventions to minimize long-term impacts. The platform will be tested in simulations and pilot scenarios to ensure it delivers clear, rapid, and equitable recommendations for decision-makers. The novelty lies in uniting an interdisciplinary team through all wildfire phases into a continuously updating, end-to-end system that links climate-aware forecasts directly to operational and health-focused decisions. Existing practices often operate in silos; no operational framework in Canada currently integrates prevention, suppression, evacuation, and recovery under a single adaptive system. The risks are significant: fusing diverse data streams, modelling rare ignition events, adapting to non-stationary climate patterns, and maintaining computational tractability under emergency conditions all push the limits of current science and technology. The rewards are equally high: reduced burned area, lower health burdens from smoke, faster and safer evacuations, and quicker recovery for affected communities, while creating a transferable architecture for other climate-driven disasters worldwide.

Sleep is a cornerstone of healthy brain development. Unfortunately, sleep disruption in neurodivergent children, including those with autism and attention-deficit/hyperactivity disorder (ADHD), is a silent epidemic—affecting up to 80%, far beyond neurotypical children. Sleep disturbances can lead to cognitive and behavioral conditions, cause caregiver stress, and challenge their academic achievement and quality of life. Current clinical systems for sleep diagnosis are too cumbersome for neurodivergent children, and wearable technologies do not consider special needs of this population. User-centered bold innovations are required to bring a paradigm shift in pediatric sleep science, transform trajectories for millions of children, and redefine brain health.

Our vision is to transform sleep care for neurodivergent youth. We will co-develop and validate StarSleep by integrating advanced sensors and physiology-driven artificial intelligence (AI) models into smart textiles; which will seamlessly, fashionably, and comfortably embed into sleepwear, bedsheets, pillow covers or socks. StarSleep will fundamentally revolutionize sleep health management and equitable access to care, in a personalized, timely, user-centred, and uninterrupted manner.

This interdisciplinary approach combines user-centered design, pediatric sleep medicine, biomedical engineering, smart textiles, and AI—fields that rarely intersect at this depth. The high-risk element lies in solving fundamental challenges in smart textiles, pediatric digital technologies, AI development for complex and noisy signals, and human factors design for children with sensory processing differences, to optimize comfort and accuracy. The high reward potential is transformative: a scalable, low-burden smart garment that can (1) detect undiagnosed sleep disorders earlier, (2) improve nightly sleep quality and daytime functioning in neurodivergent children, (3) reduce caregiver stress, and (4) inform precision medicine approaches to child development.

Beyond neurodivergent children, StarSleep can be adapted for other pediatric sleep disorders, positioning Canada as a leader in digital health innovation for vulnerable populations. By merging pediatric sleep medicine, biomedical engineering, AI-driven signal processing, and inclusive design, this interdisciplinary project will generate transformative knowledge and technology, with lasting impact on child health, family well-being, and healthcare delivery worldwide.

MOTIVATION: Cells organize some proteins into dynamic phase-separated droplets. Such droplets can produce misfolded protein aggregates leading to neurodegeneration. Molecular chaperones that promote proper protein folding interact in droplets to regulate and possibly disassemble misfolded structures, but a detailed picture of how proteins interact with chaperones in droplets and how these interactions might be harnessed to treat/prevent disease is still lacking due to shortcomings in standard approaches. Here we address this critical knowledge gap by combining disparate methods from molecular biology, chemistry, and physics. We focus on TDP-43, a protein linked to several diseases.

AIMS: 1) Catalog folding/misfolding dynamics of isolated TDP-43 molecules in droplets. 2) Quantify effects of chaperones on TDP-43 structure/dynamics in droplets. 3) Use chaperones to deliver a droplet disruptor that inhibits misfolding.

APPROACH: Aim 1: After identifying the variables defining slow vs fast TDP-43 droplet formation and aggregation, by screening disease-linked mutations and solution conditions biochemically, we will apply suitable conditions to measure folding/misfolding dynamics of single TDP-43 molecules held between two laser traps and positioned inside a droplet held by a third trap. We will repeatedly unfold and refold TDP-43 to probe what structures form, their stability and kinetics, and how behaviour varies with droplet properties (e.g. maturation), as a baseline in the absence of chaperones.

Aim 2: After quantifying how chaperone/TDP-43 interactions in droplets affect aggregation and droplet formation/maturation rates, we will tether the most effective chaperone to TDP-43 held by laser tweezers in a droplet and assess how it changes structures, folding/misfolding stability, and kinetics. Cross-linking, FRET, and mass photometry assays of chaperone/TDP-43 interactions will help build a more complete mechanistic picture.

Aim 3: We will add a bifunctional molecule that binds chaperones on one side and disrupts droplets on the other to the assays in Aims 1 & 2, to test a new therapeutic approach harnessing chaperones to deliver drugs for inhibiting misfolding.

NOVELTY/SIGNIFICANCE: This work will yield unprecedented molecular details of how protein misfolding is modulated by chaperones in droplets and test if chaperones can be exploited to deliver drugs preventing misfolding, a new strategy potentially relevant for treating many important diseases.

Canada’s health systems remain shaped by colonial legacies that marginalize First Nations voices and perpetuate inequities in access, representation, and culturally grounded care. The Truth and Reconciliation Commission’s Call to Action 23 calls for increasing Indigenous professionals in health care, yet First Nations youth remain underrepresented in health sciences due to systemic inequities, Western-centric pedagogies, and a lack of programs that affirm Indigenous identity. At the same time, First Nations communities are leading transformative, land-based, and kinship-rooted initiatives that present alternative paradigms for health and well-being.

The Land to Lab Health Sciences Summer Academy will offer a culturally grounded, land-based, and arts-integrated program for First Nations high school students from Treaty 6, Treaty 8, and the Northwest Territories. Youth will be positioned as knowledge holders, co-researchers, and leaders, blending ceremony, land-based learning, and hands-on biomedical and clinical experiences at the University of Alberta. Participants will share insights through storytelling, visual arts, and film, recognizing these as valid and rigorous forms of knowledge production.

The Academy will be led by an interdisciplinary team of First Nations scholars from the University of Alberta whose expertise spans land-based education, arts-based practice, clinical health sciences, and community-based research. Grounded in Indigenous research methodologies, the program will be anchored in ceremony, ethical spaces of engagement, and relational accountability. A Community Advisory Circle, comprising Elders, youth, Indigenous health professionals, and educators, will guide curriculum, recruitment, evaluation, and knowledge mobilization.

Objectives include: (1) Co-develop culturally centred, land-based health sciences education integrating Indigenous and Western biomedical knowledge systems; (2)Create safe, inclusive learning environments that affirm identity, strengthen cultural pride, and connect youth with role models; (3) Assess impact on identity, confidence, and career interest through participatory, mixed-method evaluation, including ACHWM©, wisdom circles, and arts-based approaches.

High risk and high reward, the Academy will disrupt traditional health sciences education, legitimize Indigenous knowledge production, and affirm that First Nations youth are not only welcome in health sciences but essential to shaping its future.

The objectives of this proposed research is to develop a system for neurointegration that avoids some of the primary technical challenges with current systems - namely the tendency of the body to react to the damage caused by implantation by scaring at the site of the electrodes diminishes the effectiveness of the electrical connection between implanted electrodes and the soft tissues of the body. Neuroprosthetics are an extremely promising mechanism to bypass nerve damage in the brain, spinal cord, or peripheral nervous system, but the extreme mechanical mismatch between typical electrode materials like platinum on Kapton films, and the ultrasoft tissues of the human body is hypothesized to create enough relative motion during regular everyday activities that scar tissues are produced. If electrodes can be made with similar or softer materials than human tissues, it is hoped that this long-term reduction in performance can be eliminated. Two major thrusts will be attempted - a liquid conductor based approach within a gel like biocompatible polymer that can be inserted like a regular wire, but be as soft or softer than human issue once implanted in the body, and a second approach using ultrafine platinum or other biocompatible electrode wires pre-formed into extendable springs and bonded within a gradient hydrogel matrix with outer modulus matched to the expected tissue stiffness. Both approaches will be applied within a phantom initially in this project to compare with current state-of-the-art electrode systems for electrical performance. Novel systems for insertion of ultra-soft electrodes will be developed as well and the most promising system combination will be used in animal models for implantation and compared for long-term performance with connection to the peripheral nervous system. If successful, this project could lay the groundwork for next generation neuroprosthetics that can be used to bypass spinal cord injury or peripheral nerve damage, or interface with the brain to treat epilepsy, recover from strokes, or improve brain-computer interfaces.

Dental fillings typically fail within 5 years due to bacterial infiltration, leading to premature failure, pain, and costly replacements, ultimately leading to tooth fractures. Current systems attempting to tackle this problem are heavily limited because they only focus on surface modifications, where if the surface is degraded the system loses all integrity or require leaching of antimicrobial materials, greatly compromising the strength of the filling and the tooth. Polymeric fillings also leach bisphenol-A (BPA), a chemical linked to serious health issues. What if we could design a non-BPA leaching dental filling that produces high-intensity antibacterial blue light on demand to kill bacteria without invasive treatment? Our project proposes a paradigm shift in dental biomaterials by developing a novel, smart dental polymer filling that actively creates an antimicrobial environment. We will develop a novel dental polymer platform based on mechanophore dioxetanes and mechanophore-piezoelectric dipeptides mixed with urethane dimethacrylate resins and test the hypothesis that oral function and ultrasound application will result in potent bacterial killing by blue light emission and electric potential generation. This work requires continuous, cross-disciplinary feedback loop, where materials science, optical physics, material characterization, and bioelectronics are simultaneously fine-tuned to test the microbiological, physicomechanical properties and biocompatibility to obtain our resultant product. The need is urgent and the potential impact of this project is enormous. More than 90% adults have decay (cavities) in their teeth, which need filling. More than 122 million polymer-based dental fillings are placed each year in North America alone. The global market size for dental filling materials is estimated to be around $20.2 billion in 2023 and expected to grow to $36.2 billion by 2030.Currently these fillings are composed of mercury-containing metal alloys (dental amalgam) or bisphenol A-glycidyl methacrylate (Bis-GMA)-based polymers. Dental amalgams are being phased out following the Minamata convention due to mercury being included in the composition, while polymer-based fillings fail often due to secondary decay (bacterial biofilms). This project will result in a principle-based, highly effective, translationally viable, and transformative dental filling material with enhanced functionality and a significantly greater lifespan than current materials.

Childhood oral disease is a critical public health challenge in Canada: 57% of children aged 6-11 are infected with Early Childhood Caries & current treatment approaches fail in 79% of cases. Despite decades of research, the mechanisms by which disease-causing oral bacteria spread within families remain poorly understood, limiting our ability to develop effective prevention strategies.

Objectives: This research aims to fundamentally transform our understanding of intrafamilial microbial transmission by quantifying, for the first time, how the contribution of family members to children's oral bacterial colonization occurs through both direct contact & airborne pathways during daily interactions. We will determine whether comprehensive bioburden reduction can break the intergenerational transmission cycles of diseases.

Method: Our interdisciplinary approach uses Ultrawide Band (UWB) spatial tracking — marking its first use in dental research — with microbial genomics & machine learning. UWB devices will monitor real-time proximity between family members with a precision of 2.1 cm. We will track the reconstitution of the oral bacterial community following professional dental therapy through sampling. Families will be randomized to receive treatment targeting different family-based bioburdens. Family contact data, air, & microbial DNA samples will be used to identify sources & mechanisms of external bacterial transmission during oral microbial recolonization.

Significance: This high-risk project challenges fundamental treatment paradigms by examining family transmission as airborne components, reframing oral pathogens as environmental contaminants that require household-level intervention rather than individual clinical management. We propose an interdisciplinary investigation across areas such as environmental chemistry, microbiology, bioinformatics, & clinical dentistry to decouple airborne versus contact transmission mechanisms & internal versus external sources of microbial reconstitution.

Impact: This study will revolutionize our fundamental understanding of oral disease: 1. Etiology: Environmental transmission within households may override genetic predisposition, challenging decades of research attributing familial disease primarily to shared genetics. 2. Clinical: It will transform pediatric dental care from individual risk assessment to family-centered environmental intervention strategies, leading to more effective therapies & reduced relapse.

Thermodynamic conversion limit in chemical processes is governed by the medium in which the transformation occurs. Recent research in chemical sciences has been focused on testing conventional organic solvents or "designer solvents" like ionic liquids to alter reaction thermodynamics. However, none of these solvents have been able to increase the thermodynamic limit of key CO2 reduction and hydrogen generation reactions. Hence, this proposal aims to use molten fluoride salt mixtures, a solvent type often used as a coolant and tritium breeder in nuclear technology, as a potential solvent for the thermochemical transformations in sustainable chemical processes. The potential high solubility of these solvents for hydrogen, as a function of temperature, will be exploited in this work to alter reaction thermodynamics. The key reaction to be taken as a case study here is the coupling of methane in the absence of oxygen (non-oxidative coupling). This transformation is of great significance to hydrogen economy since this is the only thermochemical pathway to produce hydrogen from natural gas without releasing CO/CO2 or solid carbon as a by-product. This reaction also produces ethylene, which has a market of ~200 billion $. This proposal will implement quantum mechanical based simulation tools, using Digital Research Alliance's super-computing clusters, to screen and select molten fluoride salt mixtures that have the desired thermophysical, conductive and hydrogen solubility characteristics. The selected salt mixtures will then be tested experimentally in the high temperature, molten salt reactors which are attached with the state of the art analytical system which is capable of measuring solute concentrations in molten salt mixtures at temperatures up to 1500 degrees. The proposed project aims to identify mixtures which have enough solubility of hydrogen so that the thermodynamic equilibrium limit of non-oxidative coupling of methane to produce ethylene and hydrogen increases by at least an order of magnitude, with respect to the state of the art. With worldwide market estimated to be more than 2.5 trillion $ per year by 2050, hydrogen can be the next generation energy export that can fuel jobs, economy and opportunities in Canada. The proposed project will implement the most unique interdisciplinary approach of combining chemical engineering and quantum mechanics with nuclear technology, which, to the best our knowledge, has never been done before.

Objectives

This project will design and develop an integrated prediction framework that combines numerical models, artificial intelligence (AI), spaceborne remote sensing, and Indigenous Peoples and Local Community (IPLC) knowledge to forecast critical climate-driven changes in Canada’s Arctic. Goals are to: (1) co-develop community-led monitoring of land, water, ice, and weather; (2) build the Integrated Arctic Monitoring and Prediction System (IAMPS), coupling hydrological and ocean models with AI and satellite data; and (3) predict ecosystem tipping points and thresholds of resilience to support policy recommendations that strengthen community adaptation and inform Canada’s Arctic security and sovereignty.

Summary of Approach

IAMPS will integrate physically based hydrological (Arctic HYPE) and oceanographic (NEMO) models with near-real-time AI informed by new Canadian Arctic Observing Mission data. The project addresses critical data gaps by establishing community-based monitoring networks, training local observers, and embedding IPLC knowledge to guide model evaluation and relevance. Workshops with Arctic communities in Canada, Greenland, and northern Europe revealed priority knowledge gaps around water cycle changes, river ice, streamflow extremes, and impacts on livelihoods. Predictions will be validated using both ground data and Traditional Knowledge to ensure outputs meet local adaptation needs. Results will be designed to inform governance frameworks and policy dialogues that support resilient ecosystems, sustainable communities and resources, and Canada’s sovereignty.

Novelty

IAMPS will be the first coupled hydrological and ocean model system supported by AI-driven, near-real-time predictions of the Arctic. By braiding experts and advances in physical science, political science, and social science with IPLC expertise, the project produces actionable knowledge for both communities and decision-makers. Though at high-risk of rejection by community members, IAMPS has the potential to provide a common path forward for ILPCs and trans-Arctic governments by offering unprecedented capacity to build science-driven policy for Canadian Arctic by uniting complex models, spaceborne data, and community-led observations. This research directly supports Canada’s climate adaptation policy and Arctic sovereignty due to the threat of an ice-free Arctic Ocean, and meets commitments to the UN Decades of Ocean Science, Cryosphere and Rights of Indigenous Peoples.

This project addresses a critical and underexplored challenge in heritage preservation: how to ensure digital representations of cultural heritage remain intelligible and accessible over extremely long timeframes—100 to 10,000 years. While digital twins – virtual replicas created using photogrammetry and LiDAR - have become standard tools for documenting at-risk heritage sites, their long-term preservation is threatened by bit rot, technological obsolescence, and the loss of contextual information. In response, this project introduces reconstructive preservation—a novel approach that treats digital objects not as static archives but as artifacts to be reinterpreted and reconstructed by future generations, even if degraded.

The project integrates three interrelated strategies. First, the team will develop new, purposely-designed reversible digital formats that support partial reconstruction even when files are damaged. These formats will embed resilience features such as error detection, structural redundancy, and contextual cues. Second, drawing on fields like nuclear semiotics, the team will co-create symbolic systems with Indigenous and institutional partners to embed visual markers and metadata into both digital and physical heritage representations. These systems aim to communicate the identity and meaning of heritage objects across vast cultural and temporal distances. Third, the project will explore physicalization—the transformation of digital twins into 3D-printed models that can be curated by communities with limited digital infrastructure. These physical surrogates, enhanced with embedded symbols and metadata, offer a tangible and durable alternative to purely digital preservation.

Evaluation will take two forms: forward-looking expert assessment (via Delphi consensus and predictive modeling) to anticipate how future users might interpret the project’s outputs, and backward-looking stress tests that simulate file degradation and assess reconstructability. The project is grounded in ethical collaboration and data sovereignty, especially in the context of Indigenous heritage, and builds on the team’s long-standing partnerships and extensive experience in digital documentation and community-engaged research.

By combining speculative design, advanced computing, Indigenous data governance, and digital humanities, this transdisciplinary initiative exemplifies the high-risk, high-reward mandate of the NFRF Exploration stream.

Background:

Modern electrical grids are shifting from conventional resources to power electronics (PE)-based renewables, adding high-speed dynamics that strain industry standard component-based simulation and analysis techniques. The recent massive blackouts in Portugal and Spain highlighted the need for high-fidelity analysis tools in renewable-penetrated systems.

Objective and Research Approach:

The objective of this research is to introduce a novel modeling paradigm for power systems by posing it as a spacetime continuum. This allows for applications of numerical techniques such as adaptive mesh refinement (AMR) to discretize power system partial differential equations (PDEs). These techniques have been mainly developed in astrophysics for managing nonlinear wave propagation and multiscale interactions in the most demanding spacetime simulations. The continuum formulation will propagate disturbances through spacetime like waves, allowing mesh refinement near PE-penetrated regions and coarser resolution elsewhere.

Novelty and Expected Significance:

This is a high-risk research given that the proposed power system modeling paradigm is a radical shift from the standard modeling approach. The proposed approach is untested and unlikely to receive cash or in-kind funding without a prototype. The interdisciplinary nature of the research merges two divisions within the Canadian Research and Development Classification, i.e., “Natural Sciences” and “Engineering and Technology.” Furthermore, the research cross-pollinates technologies across two disciplines that have evolved largely independently, i.e., astrophysics and power systems. The potential reward is transformative: a new modeling paradigm capable of simulating complex PE-dominated grids in real-time, thus enabling new insight into wave propagation, faster contingency analysis, and scalable stability assessment under growing grid complexity. Feasibility stems from mature AMR methods.

Artificial Intelligence (AI) can facilitate diagnostic, prognostic and clinical management decision-making in emergency settings. However, AI-based Clinical Decision Support Systems (CDSS) have had variable uptake in emergency clinical settings, with clinicians reporting increased computer time, difficulty of use, and variability in trust. We propose a high-risk, high-reward hybrid AI solution: an AI Scribe for automated risk assessment and a multimodal AI model for rapid, clinical decision support (CDS), leveraging a transdisciplinary team to transform AI-based CDS in emergency care.

Current clinical decision scoring tools, such as the ten-item Wells' Criteria for deep venous thrombosis (DVT), have demonstrated effectiveness for risk stratification for diagnostic likelihood of DVT in emergency settings. In our index example, we propose 1) A natural language processing (NLP)-based AI Scribe to automatically capture vital information from conversations during patient triaging, input from electronic health records (EHR), documented physical exam findings and clinician input to facilitate Wells' score generation and subsequently evidence-based CDS. 2) Multimodal AI to integrate inputs such as EHR notes, ultrasound imaging, genomics, and lab data (e.g. D-dimer) for precise clinical risk prediction. 3) Knowledge Graphs to provide explainable AI to deliver transparent, clinically relevant risk assessments, enhancing clinician trust.

This high-risk approach pioneers a hybrid, multimodal data integration and real-time diagnostics, with high-reward potential to revolutionize emergency care. Once developed, we can scale this design to adapt to other diagnostic, prognostic and clinical management CDS tools. Our transdisciplinary team comprises expertise in emergency medicine, internal medicine, AI and NLP, software engineering, healthcare operations, and regulatory compliance. Our collaboration with frontline clinicians ensures seamless integration into emergency workflows, supported by regulatory compliance with Health Canada and ethics boards.

Our hybrid AI solution streamlines low-risk screening, reduces diagnostic delays, imaging referrals, and costs, and improves outcomes, especially for cancer patients. AI ethics and equity expertise ensure fair predictions across diverse populations. Pilot studies will validate its impact in Calgary ERs, driving patient-centred care and potentially influencing healthcare in Canada and beyond through scalable innovation.

Wildfire smoke is an escalating global health threat, yet its full chemical and biological complexity, and the mechanisms by which it provokes steroid-refractory flares of asthma, COPD, cystic fibrosis and bronchiectasis, remain poorly understood. This project will create the first integrated “chemical-microbiome atlas” of wildfire smoke by defining the pyroaeromicrobiome, the viable bacterial and fungal communities aerosolised during controlled combustion of representative boreal fuels and captured from ambient wildfire plumes and mapping the paired aerosolome, an untargeted chemical fingerprint of the same particulate matter obtained through state-of-the-art two-dimensional gas-chromatography/mass-spectrometry and high-resolution mass-spectrometry platforms. Smoke particles collected on quartz filters will undergo microbial load quantification by droplet-digital PCR, followed by long-read metagenomic sequencing to resolve taxonomic and functional profiles; parallel chemical extracts will be analysed for thousands of semi-volatile compounds, with both datasets integrated through multivariate network approaches to reveal fuel-, region- and burn-specific exposure clusters. The high-risk, high-reward novelty lies in treating wildfire smoke as a unified biological–chemical exposure rather than isolated particulate mass, fusing aerosolomics, microbiology, atmospheric science and respiratory medicine in a unique approach that meets the mandate of the NFRF–Exploration grant. Expected outcomes include an open-access exposure atlas, biomarker panels that predict airway inflammation, and a knowledge base to leverage for future initiatives including informing realtime public-health alerts, targeted use of existing biologics and management strategies that minimize the most harmful emissions. By enabling precision intervention during smoke episodes, this work aims to ultimately be interdisciplinary through a comprehensive mechanistic understanding that in turn will translate to improvements in humans exposed to wildfire smoke through directed approaches that will reduce hospitalisations and inequitable health burdens in those individuals and communities disproportionately affected by wildfire smoke, ultimately transforming airway medicine.

The loss of myelin is common in multiple sclerosis (MS) and other neurological conditions. Remyelination is an important therapeutic strategy since it restores function. Potential remyelinating agents should best be targeted into the central nervous system (CNS) to avoid systemic toxicity but viable drug-delivery systems that efficiently cross the blood-brain barrier (BBB) are lacking. We propose a High Risk, High Reward approach to develop a novel CNS-targeting polymer platform for efficient drug delivery. These polymers of different pre-built alpha-lipoic acid - drug monomers can be homopolymers of one drug or heteropolymers that contain different drugs with complementary bioactivity. To ensure high water solubility for intranasal administration that improves CNS access, tetraethylene glycol containing a terminal carboxylic acid will be used as a linker to bridge a therapeutic drug and alpha-lipoic acid via an enzyme cleavable ester functionality. The unique advantage of our platform is the S-S disulfide-embedded five membered ring which can be efficiently broken up under very mild condition via a ring-opening polymerization process, to form linear polymers that contain a new type of S-S disulfide linkages. The resulting polymers can be reduced back in lesion site with the help of endogenous bioreductants such as glutathione to reestablish the initial alpha-lipoic acid - drug monomer and the drug can then be released by non-specific esterases that populate lesion sites. To increase lesion-homing capability of the polymers, we will introduce a CAQK tetrapeptide. These chemistry-driven high risk novel innovations will then be applied across discipline into a neuroscience-based platform to determine if high reward remyelination can be achieved in mice after an MS-like CNS injury. Here, another high risk approach is to incorporate within the polymer backbone repurposed or novel drugs from the chemistry pipeline that can effectively promote remyelination. Overall, this interdisciplinary platform to synthesize (Ling, chemist), test (Yong, neuroscientist) and deliver unique drugs selectively into CNS lesions to promote repair is high risk because it has never been achieved before. It requires both NSERC-stream (Ling) and CIHR (Yong) specialized expertise, and is impossible with either pillar alone. The program is high reward because of the potential to promote CNS repair and regain functions with minimal peripheral side effects in neurological disorders.

Chronic pain care is failing too many people, and limits of current analgesics create urgent need for alternative non-opioid therapies. Women carry a higher burden of many pain conditions. This project takes a high-risk, high-reward swing at a non-opioid solution. Instead of plugging the pore of NaV1.7, a sodium channel that drives nociceptor excitability, we will flip the channel’s internal control switches. Success opens a path to relief that preserves normal sensation; failure still delivers a map of channel regulation.

Objectives build on discovery of a site where calmodulin and 14-3-3 dock onto NaV1.7 and alter its behavior. We will identify compact, druggable contact points within a flexible intracellular region that current structures miss; create peptides or small molecules to break harmful interactions; and prove benefit and safety in cells and in vivo, with analyses powered to detect sex differences and to inform better patient care. We will also define how these interactions reshape channel trafficking to and from the plasma membrane, controlling surface density and nociceptor excitability.

The approach links atomic detail to animal behavior. First, we will measure and map binding between NaV1.7 and its regulators under different calcium and phosphorylation conditions, and obtain structural snapshots to pinpoint intervention sites. Second, we will test engineered channel variants and rationally designed disruptors in human cell models and mouse sensory neurons, using whole-cell recordings to measure changes in gating and surface expression. Third, we will evaluate the best leads in both sexes using mouse pain models: a nerve-injury model of neuropathic pain (spared-nerve injury) and an inflammatory model. Outcomes include mechanical sensitivity (von Frey), heat sensitivity (Hargreaves), and cold allodynia (acetone or cold plate), with safety and locomotion checks.

The novelty is turning a understudied flexible regulatory region, long ignored by drug discovery, into a precision target. By reprogramming the channel’s control system rather than blocking ion flow, this strategy could normalize overactive nociceptors without numbing healthy sensation. Expected significance: a first-in-class, structure-guided route to non-opioid analgesia that addresses inequitable pain outcomes, plus structural templates, quantitative maps, and in vivo proof-of-concept that reset how the community targets ion channels in pain.

The sense of touch is essential for motor tasks and for interacting with social and physical environments. Tactile acuity varies across species, with humans possessing highly sensitive, dexterous hands. Touch sensitivity also changes with age, typically declining over time. Manual touch depends on sensory neurons that innervate the skin, yet how they function and vary within and across species remains poorly understood. This gap limits our understanding of touch evolution and ontogeny.

While much research has focused on central processing of touch signals, recent studies show that tactile neurons themselves engage in complex processing through branched axons. These neurons, which innervate Merkel cells and Meissner corpuscles, have receptive fields with multiple sensitive zones or “hotspots.” As an object moves across the skin, these hotspots are sequentially activated, producing precise spatiotemporal patterns of action potentials that encode tactile information.

We hypothesize that the axon branching structure of first-order tactile neurons influences manual dexterity. For instance, capuchin monkeys show much higher dexterity than howler monkeys. We will compare axon branching across monkey species and extend this analysis to rodents, using mice for physiological experiments. These data will support computational modeling to interpret cross-species variation.

Aim 1 – Quantify axon branching complexity in forelimb somatosensory neurons of three monkey species (capuchin, howler, macaque). We hypothesize that branching will be most complex in capuchins and least in howlers. Axon branching will also be assessed in mice and rats. In macaques and mice, samples will span different ages to assess age-related changes.

Aim 2 – Determine how axon branching complexity affects tactile encoding. Using transgenic mice expressing a voltage-sensitive indicator, we will image axon terminal structure and function. These data will inform computational models to examine how branching complexity shapes tactile encoding.

This interdisciplinary project brings together Amanda Melin (primate sensory systems), Steve Prescott (somatosensory neurophysiology and modeling), and experts Drs. Ratté and Pruszynski, who contribute expertise in neuron morphology and primate somatosensory physiology.

In Canada, Indigenous communities are impacted by environmental risk from industry and changing climate, posing threats to environmental and community health. Current environmental management and risk assessment models fall short in responding to critical needs of affected communities. At the request of Oneida Nation of the Thames (ONT) and Sipekne’katik First Nation (SFN), this project seeks to establish a scalable place-based methodological framework for water management that considers coupled ground- and surface water spanning various spatiotemporal scales.

Objectives:

Develop culturally relevant conceptual frameworks for the design and implementation of hydrological models that incorporate community conceptions of risk and wellbeing Collect site-specific environmental data to identify dynamic groundwater flow pathways connecting contaminant sources to receptors

Characterize environmental risk and identify mitigation and remediation strategies

Contribute to capacity building with community partners through knowledge translation (e.g. observation and monitoring data, training opportunities, citizen-science data collection)

Support community-driven knowledge mobilization (e.g. science-policy integration, advocacy, art)

As an Indigenous-led, interdisciplinary research group (MG360 and IndigenERA) with expertise in Indigenous environmental health and groundwater research, the research team is uniquely positioned to address existing research gaps. This research embodies a high-risk, high-reward approach by directly challenging conventional water management risk assessment methodologies that disadvantage Indigenous perspectives and priorities and disregard dynamic groundwater flow networks. Current community risk burdens, demand new integrated systems-based frameworks. Our project addresses the challenge to develop a community-led, culturally grounded approach to water modeling, which departs from mainstream science and risk assessment models. The inherent complexities of conducting research with communities (not on them) are addressed through co-designed engagement frameworks that prioritize relational accountability and mutual learning. The potential rewards are transformative: Supporting Indigenous sovereignty over health with broad impacts across scientific innovation, economic costs of health and remediation, and cultural revitalization. Crucially, this project supports policy advocacy by providing tools and data to advocate for pollution remediation

Canada’s agri-food sector demand for food safety compliance is at its highest, yet the application of digital models, ghosts/shadows or twins to microorganisms remains underexplored. Salmonella is one of the most prevalent foodborne pathogens worldwide, contributing significantly to food-related illness and economic losses. Despite stringent safety protocols, outbreaks continue to occur due to limitations in real-time monitoring, prediction, and control strategies. Addressing this issue requires a transformative approach to understanding and managing pathogen behavior across complex food production and distribution systems.

This highly interdisciplinary project will integrate approaches from microbiology, chemistry, systems biology, food processing, data science, and computational modeling to develop digital models/twins of Salmonella—virtual representations of the pathogen that simulate its behavior and interactions in dynamic food environments. These digital twins will enable real-time forecasting of contamination risk by modeling the microbial response to various environmental, processing, and storage conditions. Central to this development will be the use of high-throughput genomic and physiological data and environmental parameters, which, coupled with advanced computational modeling, will simulate microbial behavior under varying environmental and metabolic conditions.

These digital models, with enhanced predictive ability compared to extant approaches, will improve real-time monitoring, control, and optimization of risk reduction efforts in the food industry. A key application will be the use of these digital twins to assess pathogen survival and growth in perishable foods, with a focus on ready-to-eat products. We aim to estimate contamination likelihood, pathogen load, and optimal intervention strategies in real time. This will not only enhance food safety but also reduce unnecessary food waste due to overly conservative shelf-life estimations.

The project pioneers the convergence of computational and biological sciences in a novel way, tackling a multifaceted global issue through the lens of pathogen modeling. Although high-risk due to the complexity of microbial systems and variability in real-world conditions, the potential impact is enormous—revolutionizing food safety management and offering a powerful tool for protecting public health and ensuring sustainable food systems.

Emerging studies suggest that electromagnetic therapies targeting cognitive decline in dementia may also influence vestibular function, though their underlying mechanisms of action remain unclear. Animal data show the vestibular system is sensitive even to weak magnetic fields-such as those generated by the Earth’s geomagnetic field-indicating that magnetic simulation may indeed modulate vestibular activity.

This project investigates whether low-intensity dynamic magnetic stimulation can induce a measurable vestibular afferent response. Specifically, we ask: Can low-level magnetic stimulation induce a detectable change in vestibular afferent activity?

This research's objectives are to

1) evaluate afferent responses to the magnetic stimulus,

2) quantify resulting changes using Electrovestibulography (EVestG), and

3) compare these effects with those from visual and whole-body tilt stimuli.

Traditional assessments of vestibular function often rely on indirect, stimulus-evoked measures. EVestG is a notable exception, offering a more direct method. EVestG typically uses whole-body tilts to assess vestibular performance, which has been shown to decline in various neurodegenerative and neuropsychiatric conditions. A key advancement would be a stimulator capable of inducing a subtle threshold shift in vestibular activity. Studies of the visuo-vestibular connection show that visual stimuli can modulate spontaneous vestibular afferent firing via the Efferent Vestibular System (EVS), primarily through threshold-shifting mechanisms.

We propose that a magnetic stimulus, if capable of inducing such a shift, could generate a measurable EVestG response. This project focuses on developing and validating such a device. This project departs from conventional approaches by aiming not to evoke a driven vestibular response, but rather to induce a threshold shift. Since the Efferent Vestibular System (EVS) primarily acts by modulating afferent thresholds—as shown in visuo-vestibular stimulation studies measured by EVestG—we aim to develop a low-level magnetic stimulator to elicit a measurable threshold shift in vestibular afferents.

This novel stimulation strategy could offer a new tool for assessing vestibular function and monitoring changes linked to neurophysiological and neurodegenerative conditions. Crucially, it opens the door to portable, dynamic EVestG assessments, potentially enabling home-based monitoring.

Biomaterials hold tremendous potential to prevent the numerous cases of female infertility and pregnancy-related complications associated with uterine scarring resulting from common surgical procedures (e.g. Cesarian section). This is particularly crucial for patients in remote and low-resource settings, where access to comprehensive post-surgical care is often limited or altogether lacking. To promote scar-free uterine tissue regeneration and prevent associated post-operative complications (e.g. intrauterine adhesions, placenta accreta), we propose a novel biomaterial that is specifically designed as a minimally invasive wound-healing solution. Our technology distinctively integrates clinically essential features into a single, low-cost system: (i) it is bioinspired and sustainable, derived entirely from food- or agricultural-grade materials; (ii) it is injectable and undergoes in situ gelation at body temperature; (iii) it provides antimicrobial and anti-inflammatory functions. Importantly, (iv) the formulation is cost-effective and field-deployable, requiring no refrigeration or specialized handling, making it uniquely suited for scalable use in maternal care across remoted and resource-limited communities.

To this end, we will synergistically integrate the distinctive properties of naturally derived polymers and compounds (e.g. gelatin, curcumin) to support the scar-free regeneration of endometrial and myometrial tissues. Experimentally, we will first ensure that target performance metrics (e.g. injectability, degradation) are met. Subsequently, the biomaterial’s biological performance will be validated within both an organoid system and a mouse model of uterine injury.

The high-risk nature is comprised by the objective to capitalize on a completely new approach to treat uterine injuries, aiming at minimizing pregnancy-related complications and surgery-related infertility. Secondly, this research intends to radically challenge long-standing practices with a strategy that uniquely addresses both immediate needs and detrimental long-term consequences. The high-reward nature is comprised by the promise to enable significant societal, technological and health impact. In fact, this novel biomaterial is expected to reduce healing time and prevent post-operative complications. Lastly, our work will lead to ground-breaking advances in the development of a bioactive materials, ultimately promising a quick translation of fundamental discoveries to the clinic.

Early detection is critical to improving cancer outcomes and our proposed technology seeks to identify malignancy at the molecular level, before clinical symptoms emerge. We will develop an innovative spectroscopic imaging platform leveraging terahertz (THz) radiation to probe molecular vibrations and dynamic structural changes in cancerous tissues.

Current diagnostic methods are limited in their ability to detect rapid, subtle, and often irreversible molecular changes in biological samples, changes that may signal early disease onset. To overcome this, we will harness cutting-edge technology that leverages far-infrared radiation in the THz frequency range to monitor vibrational resonances and dynamically observe key biochemical processes. While well-established in materials science, applying THz spectroscopy to biological systems is largely uncharted and presents a high-risk, high-reward opportunity for breakthrough discovery. We propose, for the first time, to use a real-time THz monitoring system with microsecond time resolution to capture how organic materials dynamically respond to rapid structural or energetic changes. We will initially validate our concept with lysozyme, a well-established model for studying enzyme catalysis and protein dynamics, then apply it to characterize both tumor and healthy tissues from a biobank to enable comparative analysis. This will allow us to identify novel biomarkers that reflect molecular changes associated with cancer and related conditions, offering a new dimension beyond traditional static molecular profiling.

Our initial focus is endometriosis, a chronic inflammatory disease affecting approximately 10% of women of reproductive age and linked to a tenfold increased risk of developing ovarian cancer. Current diagnostic methods for endometriosis are inadequate, often resulting in delays of 5 to 10 years. Our objective is to identify dynamic spectral signatures that reveal molecular compositions and structural changes, enabling characterization of the tissue microenvironment in endometriosis and its potential progression to ovarian cancer, ultimately redefining how diseases like cancer are diagnosed and monitored.

This interdisciplinary effort merges advances in physics, chemistry, biology, and clinical research to pioneer a new diagnostic paradigm, transforming static molecular snapshots into real-time, functional insights with significant potential to improve cancer detection and patient outcomes.

The emergence of drug resistance remains a major challenge in cancer therapy. While standard chemotherapy can induce initial remission, most patients ultimately relapse and develop resistance to treatment. At the cellular level, chemotherapy imposes a hostile environment on malignant cells. In response, a subset of cells enters a dormant state, reducing their metabolic activity to survive cytotoxic stress without acquiring new genetic mutations. Once drug exposure ceases, these drug-tolerant persister (DTP) cells can resume proliferation, driving disease relapse. This adaptive behavior mirrors hibernation in animals, where species employ metabolic depression to withstand adverse conditions such as extreme temperatures or food scarcity. We hypothesize that the molecular pathways regulating animal hibernation may provide valuable insights into mechanisms of drug resistance in cancer and reveal novel therapeutic targets. In this project, we will employ artificial intelligence (AI) and bioinformatics to investigate the molecular features and pathways of hibernation. We will reconstruct the hibernation process as a dynamic network across timepoints and, through network science approaches, identify weak nodes that represent potential intervention points. By integrating AI with cancer pharmacogenomic datasets, we will prioritize drug candidates capable of targeting these network vulnerabilities. Promising compounds will then be validated through in vivo drug screening. By leveraging evolutionary parallels between animal dormancy and cancer cell persistence, this project aims to uncover innovative therapeutic strategies to overcome chemoresistance. Ultimately, our work has the potential to open new avenues for durable cancer treatment and significantly improve patient outcomes.

Engineering is historically a male-dominated field; in 2023, women were only 20.2% of licensed engineers in Canada. Lack of diverse representation in engineering has led to designs that are used by the public, e.g., car seats, that function at a lower level for females compared to males. The need to increase diverse representation in engineering is highlighted by Engineer Canada’s “30 by 30” initiative, but little progress has been made. Issues of career retention and lack of representation in role models are known factors, and recruitment is an area where development could help address representation issues. Our research aims to re-frame engineering through classroom-based activities that focus on the communication, creativity, and social benefits inherent to engineering. The goal is to empower and attract a more diverse set of students by highlighting these soft skills.

Objectives for this research project include creating activities that tie into engineering competencies and empower equity-deserving students to find a personal connection to engineering. These novel activities integrate soft skills alongside math and science to build a holistic picture of the profession. We will work with active and pre-service teachers to deliver classroom activities and will also lead professional development workshops.

Beyond improved diversity in the field of engineering, expected outputs include an adaptable career exploration classroom activity for use by teachers across the country. By presenting these activities in workplaces and to governing bodies, we also have the potential to challenge the engineering mindset to better value the social and artistic parallels that are integral to engineering design. Expansions of engineering pedagogy towards more inclusive and authentic teaching and learning can also help shape future engineers to champion inclusive and diverse workplaces. All these potential rewards carry a caveat of social buy-in from teachers, engineers, and students. Risks associated with this project include potential resentment towards a more inclusive view of engineering and for all interventions and activities to have no net effect. To date, there has been little progress towards improving diversity in engineering, despite much research. By presenting engineering as a discipline that embraces creativity and social connections, our novel educational intervention will foster a new generation of engineers capable of transforming design practices.

Tuberculosis is still the number one killer amongst infectious diseases, responsible for 10 million new infections and 1.3 million deaths every year. BCG, the only vaccine currently in use against TB offers limited and highly variable protection and is particularly poor against adult pulmonary TB. Developing an effective, fully protective TB vaccine faces several challenges – an incomplete understanding of the host-pathogen interactions, lack of reliable biomarkers or correlates of protection, heterogenous and diverse clinical states of the disease and the lack of suitable animal models. While animal models have been indispensable for evaluating the safety, immunogenicity and efficacy of TB vaccines before human trials, they fail to fully recapitulate the full spectrum of human TB and often do not replicate essential aspects of human immune responses. We propose to establish advanced lymphoid organoids and lymph-on-chip platforms to investigate Mycobacterium tuberculosis infection, host immune responses and the efficacy of novel vaccine candidates. Using the advances in biomaterials engineering and learnings from bioengineered hydrogels used in tumor biology, we will develop two primary in vitro human lymphoid models

1) Human lymphoid organoids – Derived from human tonsil/adenoid tissue or peripheral blood mononuclear cells, these 3D cultures will replicate the micro-architecture of native lymphoid tissue.

2) Lymphoid organ-on-chip – These microfluidic devices will provide a controlled environment for culturing human immune cells, enhancing physiological relevance and enabling dynamic antigen delivery and immune cell interactions.

We will use these models to elucidate Mtb pathogenesis, characterize human-specific immune responses and use this platform to test novel vaccine candidates. These models will help us investigate how Mtb establishes infection and evades immunity and to study the interactions of immune cells, including B cells and T follicular helper cells in context of Mtb infection and vaccination. This proposal represents a “human-first approach” to TB research, addressing the limitations of animal models and providing a platform for personalized medicine.

While the project faces challenges in achieving full physiological mimicry and standardization, the potential rewards are transformative. Success could revolutionize TB research, accelerate vaccine and immunotherapy development and enhance our understanding of human immunology.

We propose developing an affect-aware, generative AI tutor—Abel Tutor—that can detect and respond to cognitive and emotional states in real-time. Using only the built-in camera and microphone on standard devices, we will triangulate data from facial expression analysis (FACS Action Units), Transdermal Optical Imaging (TOI), voice dynamics, and pupil dilation to infer students’ cognitive engagement and understanding phases, without wearable sensors.

Objectives:

-Detect learning states (nascent, confusion, emergent, deep understanding) by analyzing subtle physiological and vocal cues.

-Develop machine learning models that fuse multimodal signals to predict cognitive processes.

-Develop an adaptive tutoring agent that dynamically adjusts instruction based on students’ trajectories. 

Innovative Contribution:

This research integrates cognitive psychology, machine learning, affective computing, educational theory, multi-modal perception, and astrophysical signal processing. It breaks new ground by:

-Triangulating physiological signals (e.g., facial blood flow, micro-expressions, pupil dilation) using only consumer-grade hardware.

-Modelling subjective phases of understanding—an area previously out of reach due to technical limitations.

-Enabling Gen-AI to adapt based on emotional valence and mental state, beyond static or rule-based personalization.

High Risk and Reward:

Multimodal signal fusion for cognitive detection is a challenging task and is rarely applied in educational settings. Risks include signal reliability, model error, and feedback efficacy, but success would yield a real-time AI tutor that fosters deep learning, self-regulation, and educational equity.

Interdisciplinary and Transformative:

Our team combines expertise in astronomy to enhance the quality of TOI signals using computational imaging methods, robotic engineering to advance non-verbal perception, affective reasoning, and intelligent autonomous systems, data science for multimodal signal fusion and analysis, and applied cognitive science to model learning phases and cognitive-affective dynamics. This unique collaboration bridges affective neuroscience, embodied cognition, robotics, speech science, data science, and astrodata modelling, laying the foundation for emotionally intelligent AI that holistically supports human learning. Aligned with this stream, our approach seeks to test a paradigm-shifting idea that may fail-but, if successful, would redefine the role of AI in education.

Antimicrobial susceptibility testing (AST) is a crucial clinical diagnostic that helps physicians prescribe appropriate antibiotics for treatment of infectious diseases. Standard AST methods have a turnaround time of ~16-24 hours, which forces clinicians to prescribe treatments without prior knowledge of their effectiveness. Rapid AST methods can return results in ~6-8 hours, which have been shown to significantly improve clinical outcomes. However, these methods are complex and costly, limiting their widespread use.

Objective: We aim to develop a new rapid, low-cost clinical workflow for AST that delivers results within 1–2 hours. Our approach leverages microdroplet technology to encapsulate bacteria with antibiotics and fluorescent dyes, enabling highly multiplexed real-time susceptibility profiling through automated imaging and machine learning (ML).

What we have: We have recently discovered that bacterial drug resistance phenotypes can be detected in 1-2 hours when they are encapsulated into microdroplets, which is ~10 fold faster than the current gold standard. Using this core technology, we envision a clinical workflow where bacterial samples are encapsulated with different antibiotics into microdroplets. These droplets are then loaded into an imaging chip and placed in a compact imaging system for fluorescence monitoring. ML algorithms interpret the fluorescence kinetics to infer susceptibility in real time.

What we need: While the results are robust in the lab, we do not know the extent to which these results will apply for bacteria from clinical samples. We also must test the system’s performance on the full panel of antibiotics used in clinical susceptibility testing.

What we propose: 1) build and optimize a microdroplet system to perform multiplexed AST for 10 clinically important antibiotics and 2) use this system to test a wide range of clinical isolates, comparing results to the current clinical gold standard.

High-risk aspect: The key challenge lies in achieving reliable, multiplexed detection of resistance phenotypes, across diverse clinical strains and antibiotic classes.

Potential impact: If successful, this technology would cut AST turnaround times by 4-fold, significantly improving clinical decision-making, reducing inappropriate antibiotic use, and lowering hospital costs in Canada and globally.

Dementia is a progressive syndrome marked by declines in memory, executive function, and daily functioning. Preventing or delaying progression from subtle cognitive change to mild cognitive impairment, and ultimately dementia, is a public health priority. The 2020 Lancet Commission and its 2024 update emphasize prevention and identify fourteen risk factors across the life course, including air pollution. Canadian environmental exposures are typically evaluated one at a time and mainly against dementia incidence or mortality in administrative cohorts rather than longitudinal cognitive trajectories. International studies link long-term air pollution to dementia and suggest extreme heat may accelerate cognitive decline in socially vulnerable groups; however, Canadian cognition-focused evidence on joint exposure to urban air pollution and heat remains limited.

The overarching goal is to quantify the joint dose–response relationships between urban air pollution and heat and cognitive decline in older Canadians, and to develop reproducible, reusable exposure data and spatial causal inference frameworks for broad impact. This study has two aims: i) Machine learning exposure mapping: Building on the exposure layers of the Canadian Urban Environmental Health Research Consortium, we will use state-of-the-art machine learning ensembles to estimate fine-resolution ground-level air pollution and heat by integrating multiple data sources, including built environment factors. ii) Causal spatial analysis of cognitive trajectories and social vulnerability: Using the Canadian Longitudinal Study on Aging, we will align fine-resolution exposure surfaces to residential postal codes, and quantify the causal dose–response and joint effects of air pollution and heat on repeatedly measured cognitive functions, and assess heterogeneity by socioeconomic status.

Our project is high risk. This is the first Canadian study to combine co-exposures to air pollution and heat with longitudinal cognitive assessment to estimate causal dose–response effects at a population scale. Our project is high reward. The results will inform urban heat and air quality policy and enable integration with established national research platforms. Our interdisciplinary team, strengthened by international collaborators in environmental science and spatial causal inference, is well positioned to translate findings to shape understanding of environmental determinants of brain health in aging populations.

This interdisciplinary, high-risk, high-reward project explores the use of giant puppetry as a transformative, arts-based strategy to engage Kenyan youth in climate justice advocacy. Integrating cultural arts, environmental science, and social justice, the project will create an experimental platform to disrupt conventional climate communication and reimagine public engagement.

Drawing from Kenya’s vibrant traditions of storytelling, community theatre, and puppet-making, this initiative will bring together youth collaborators, artists, environmental educators, and climate scientists. Working collectively, participants will co-design and perform with large-scale puppets that embody urgent environmental narratives—such as youth lived experiences of drought, water scarcity, gender-based violence, resilience, and ecological stewardship. These public performances in urban and peri-urban Kenyan communities will function both as acts of creative resistance and as vehicles for dialogue, activism, and learning.

Core objectives:

• Foster youth-led storytelling and design, enabling participants to create, script, and animate narratives grounded in their lived climate realities;

• Catalyze public awareness and civic discourse, leveraging puppetry’s symbolic and accessible power to reach diverse audiences;

• Test the ability of participatory performance to shift perceptions, inspire local action, and influence informal policy processes;

• Explore the approach’s replicability and scalability across other contexts facing climate injustice.

This project exemplifies the NFRF Exploration program's commitment to interdisciplinary, unconventional, and bold approaches. It merges artistic practice with environmental research and community engagement to address one of the most pressing challenges of our time. By mobilizing youth in Kenya—one of the regions most vulnerable to climate impacts—the project advances equity by amplifying marginalized youth voices in climate discourse. UNICEF's 2024 The State of the World's Children Report identified climate crises as a megatrend shaping youths' futures, yet note youths' lived experiences and climate-related priorities are understudied.

Over two years, the team will prototype, stage, and evaluate a series of puppet-based interventions. If successful, this project offers a novel, scalable model for arts-based climate justice engagement, animating a more just and imaginative future through youth creativity and collective performance.

A new class of image sensors—high-speed, free-running 2D SPADs (single-photon

avalanche diodes)—allow detecting and time-stamping individual photons with

micro- to picosecond precision. Unlike conventional sensors that integrate light,

2D SPADs allow properties of individual photons (pupil entry point, optical distortion,

2D position, time of flight, etc) to be estimated and adjusted before final image

formation. This challenges the traditional model of light integration and opens the door

to revolutionary forms of imaging where optics and integration are (partially)

replaced by processing photon data streams.

We propose to apply this approach to two high-impact settings: astronomical telescopes and

endoscopic probes.

Our first thrust will be to realize “Adaptive Optics in Silico (AO-Si)”---aimed at

developing high-resolution, wide-field terrestrial telescopes in the visible band,

which is a major frontier in astronomical instrumentation. Unlike infrared-band

counterparts that use adaptive optics (AO) refined over decades, visible-band AO faces

enormous, still-unsolved challenges due to the shorter wavelengths involved. AO-Si

sidesteps these by shifting the most difficult AO task to computation: rather than relying

on deformable mirrors to “undo” atmospheric blur in real time, the stream of time-stamped

photon detections enable wavefront sensing, aberration correction, and image formation to

be done in software after the exposure.

Our research will advance AO-Si by (1) evaluating key design trade-offs in simulation, (2)

building a lab-scale prototype, and (3) implementing the system on a telescope. This

agenda poses significant challenges, requiring novel optics, new telescope

instrumentation, and tackling telescope deployment logistics.

Our second thrust will be to realize time-of-flight 3D imaging in scanning fiber endoscope

technology (3D-SFE). This technology enables wide-field imaging inside the body with

ultra-thin probes, but their short laser dwell times prevent conventional time-of-flight

3D sensing. By exploiting advances in ultra-wideband flux estimation from photon timestamp

streams, we aim to achieve depth and intensity imaging with one ultra-thin probe.

By pursuing these two directions, our program sits at the intersection of computer vision,

astronomy, and biomedical imaging. It has the potential to transform both fields, enabling

next-generation visible-band telescopes and minimally invasive, wide-field 3D endoscopes.

Health misinformation spreads faster than expert review, especially across multilingual and False stories on social media are 70% more likely to be reshared and reach 1500 people six times faster than true stories. Additionally, exposure to vaccine misinformation lowers intent by about six percentage points. Health misinformation outpaces expert response, with greatest harm in multilingual and under-served communities. The objective of this project is to deliver an early warning system that reduces exposure to high-harm claims and saves responder time. An ancillary objective is to provide open protocols, evaluation datasets, and a provenance-aware retrieval toolkit for safe AI in health communication.

Objectives: 1) Discover emerging health claims in legal sources and link each to its provenance (e.g., origin, cross-platform mutations, synthetic media) while estimating harm and uncertainty. 2) Generate retrieval-augmented evidence cards that cite trusted sources, explain risks for different audiences, and route time-sensitive alerts. 3) Demonstrate safety and impact in two pilots, targeting a 24 to 48 hour lead time and a 30 to 50 percent exposure reduction, with attention to multilingual access, SGBA+, and equity.

Approach: PROOF-Health is a provenance-first, retrieval-only agentic platform to detect, trace, and prioritize high-harm claims without generating or amplifying content. Multimodal detectors analyze text, image, and video, while provenance signals (e.g., metadata, watermarks, cross-platform lineage) are fused. A harm and uncertainty model then drives triage and alerts. Outputs are citation-grounded, and the agent abstains when evidence is weak. An AI-hygiene layer enforces non-amplification and audits hallucination and sycophancy via contradiction tests, adversarial probes, uncertainty calibration, and human review. Methods span machine learning, human computer interaction, information science, epidemiology, philosophy of medicine, and health policy to ensure usable tools, multilingual discovery, rigorous evidence standards, and clear adoption pathways. Governance addresses ethics, privacy, transparent logging, and community oversight.

Novelty/significance: A non-generative, non-amplifying, provenance-aware agent that unifies open-world discovery, verification, and safe triage to reduce exposure to high-harm health misinformation, with a prototype, open datasets and benchmarks, and playbooks to guide Canadian and international standards.

Objectives

To combat climate change and enhance quality of life, significant efforts have focused on improving buildings, which are both major sources of greenhouse gas emissions and our primary living spaces. However, existing building energy solutions, such as thermostats, often rely on correlation-based models that struggle to adapt to changing conditions, such as seasonal shifts or evolving occupancy patterns. This is a critical issue in human–building interaction, where occupant behavior changes over time or in response to new environments.

The proposed research aims to address this challenge by developing a novel causation-based model tailored for occupant-centric building energy solutions, with the primary objective of improving predictive reliability in dynamic, real-world settings.

Research Approach

We introduce sparse causal learning, a novel paradigm for causal modeling using high-dimensional observational data. Unlike traditional techniques that rely on correlation, this framework uncovers true causal relationships between multiple high-dimensional predictors and a single outcome. It leverages sparsity in causal effects to account for unmeasured confounding without requiring additional covariates.

Pilot studies using real-world data, including the Ecobee program dataset from over 200,000 households across North America, have demonstrated that our causation-based models significantly outperform traditional models in predicting occupant behavior and energy usage.

Novelty and Expected Significance

This research represents a groundbreaking advancement in both methodology and application.

Technological Advancement: (1) A novel sparse causal learning method that effectively mitigates bias from unmeasured confounding. Designed for multi-cause settings, it leverages sparsity in causal effects and eliminates the need for extra covariates, making it highly adaptable across diverse applications. (2) Causation-based models that provide more reliable predictions, even in the face of shifting conditions such as seasonal changes or new occupancy patterns.

Environmental Impact: The research has the potential to significantly reduce greenhouse gas emissions by improving energy efficiency in buildings.

Occupant Comfort and Well-being: This project lays the foundation for future research in smart energy systems, paving the way for more sustainable and adaptive building energy solutions that better meet the evolving needs of occupants.

This interdisciplinary project partners researchers from the University of Victoria (Visual Arts & Engineering) and Camosun College (Camosun Innovates) to augment and preserve intergenerational knowledge transfer of carving practices in Northwest Coast Indigenous art, by designing and building an adaptive robotic system that operates in collaboration with a master carver.

Carving is core to Pacific Northwest Indigenous Peoples’ social, legal, and cultural orders; from ornamental to utilitarian, miniscule to monumental, carvings (canoes, drums, masks, poles rattles, structural support for bighouses, etc.) integrate into every aspect of life. Traditionally, carving is passed down through close, hands-on mentorship, often within families or communities, but current efforts face increasing disruption (climate change, cultural shifts, land/language loss, legacies of colonialism, urbanization). This project offers new ways to enhance generational continuity and help young people engage and learn through novel digital collaboration technologies that protect and strengthen carving knowledge—not automating artistic methods but complementing carvers’ tools for communicating culturally-rooted practices over time.

Integrating deeply-embodied, culturally-grounded artistic knowledge with emerging robotic technologies involves high risks (technical and conceptual) and high rewards (generational knowledge transfer, and commercial application/development). Developing a robotic system capable of real-time adaptation to carving variable, unpredictable cedar microstructures combines experiential knowledge and technical innovation. The work provides a careful, respectful approach that upholds the integrity and sovereignty of Indigenous knowledge systems while utilizing advanced computational models and processes. These new ways to access, interpret, preserve, and teach traditional knowledge ethically embed robotics into art, education, and practice. Creating such a system allows new technologies to promote, respect, and strengthen cultural resilience.

The impact of this project is enhanced by alignment with The Seedling, the Nominated PI’s federally-funded cultural, environmental, and political art project involving planting a Western Red Cedar, designing a digital 3D totem and, and carving it in 600 years when the tree is mature. The Seedling rethinks colonial concepts of economy, law, and politics, while radically expanding planning/decision-making timescales.

The future is hard to predict, and realizing the kind of future we want in the face of multiple global pressures is harder still. Coastal Indigenous communities are facing unprecedented stressors related to climate impacts, changing food systems, and other environmental changes that must be responded to, and responding to these changes introduces new changes also. The future is therefore both unknown and certainly different. Strategically navigating these changes requires locally appropriate solutions that are guided by Indigenous Nations and informed by their history and priorities. Our transdisciplinary research program will focus on collaborative, community-engaged research with coastal British Columbia (BC) communities. Together, we aim to develop adaptation plans grounded in community history, priorities, and visions for the future, and rigorously stress-tested against multiple possible contexts.

To do this research, we will collaborate with Indigenous communities to examine the past, present and future of environmental change and adaptation to those changes. First, we will investigate past environmental shifts and community responses by drawing on archaeological record and oral histories. Second, we will document current priorities and visions for the future as described by community members. Third, we will use these insights to design simulations of possible futures, integrating projected environmental change, local community concerns, and expertise from Indigenous knowledge holders and Western science experts. These simulations will be built in the form of “serious games”, whereby decision-makers explore scenarios, make choices, and experience the consequences through feedbacks. This approach allows us to test the resilience of different adaptation options across a wide range of uncertain futures.

Advances in AI and computer science, combined with long-standing relationships with Indigenous Nations creates new opportunities to evaluate, learn from, and prepare for environmental uncertainty in ways that are both locally grounded and broadly applicable. By ensuring local relevance and systematically stress-testing options, this work will inform real-world decision-making and serve as a transferable model for environmental planning in different contexts.

The study of Dark Energy — the name given to the unknown phenomenon driving the accelerated expansion of the universe — is an exciting and evolving field of physics. Over the last year, experimental results from the Dark Energy Spectroscopic Instrument (DESI) have revolutionized the field, suggesting that the dark energy is not constant as previously believed, but dynamic, with a density weakening over time. At present, these inferences have a relatively low level of significance and only use simple statistical methods to compare the data to physical models. The exciting discovery of dynamical dark energy has the potential of being confirmed or refuted by the careful matching of future empirical data to existing physical models. Such analyses will require bespoke advanced statistical methods in order to provide robust inferences about dark energy.

To realise this goal over the next two years, we will form an interdisciplinary team to adapt methodological expertise from the fields of biostatistics and computational statistics and apply the tools developed to cosmological data. Biostatistics and cosmology are both tasked with drawing robust conclusions from limited available data, and computational statistics handles the fitting of the complex statistical models required to achieve this. Bringing together experts in these disparate fields therefore offers a chance to critically look at the new cosmological data from a novel viewpoint and confirm — or overturn – the discovery of dynamical dark energy.

Our research will focus on three directions. First, we will develop methods to draw robust statistical inferences from data that combine large-scale information on dark energy with that from smaller scale, non-linear physical processes that constrain some of the model parameters. We will exploit machine learning techniques to disentangle the dependence at the model-fitting stage. Second, we will test the match between the dynamical dark energy model and the data using techniques derived from those used in biostatistics to detect whether the observations support the assumed model. Third, we will develop new techniques for analyzing cosmological data, including alternative ways of performing model selection to ensure we make efficient inferences without faulty modelling assumptions. The ultimate aim of this project is to understand what we truly know from the new data about the physics of dark energy.

The memory robbing illness Alzheimer’s disease (AD) affects 60 million people worldwide, costs billions annually, and has no cure. Finding new AD treatments is urgent as cases will triple by 2050. Weak magnetic fields can improve memory, but classical physics and biology cannot explain how the small energies involved, far below thermal noise, produce the observed effects. This prevents translation into therapeutic use. However, these effects can be explained by quantum spin dynamics (i.e. the radical pair mechanism), directly linking memory to subatomic processes below the level of neurons. This challenges current models of brain function, which start at the level of neurons, and of physics, where quantum processes are easily disrupted by the environment. Demonstrating if quantum processes, where particles exist in multiple states and interact instantly, can influence neural function and cognition is therefore a fundamental challenge requiring an integrated interdisciplinary approach. Demonstrating this would transform our understanding and treatment of the brain, while inspiring quantum brain-based models to advance artificial intelligence (AI). We have shown that weak magnetic fields alter microtubule (MT) growth consistent with a radical pair mechanism. MTs are dynamic protein filaments that support neuron structure and are critical to synaptic plasticity and memory-related processes. Our hypothesis is that MT dynamics are regulated by spin-sensitive radical pair reactions, and that these reactions respond to external magnetic fields to modulate MT growth, neuroplasticity, and memory. To test this hypothesis, we will detect, identify and manipulate radical pair spin dynamics during MT growth using electron spin resonance spectroscopy while varying magnetic field strength and frequency. To establish a link between the spin-modulated MT dynamics and structural and behavioral outcomes, we will assess changes in neuron shape in vitro and memory performance in vivo in rodent models exposed to magnetic fields. Finally, to assess how weak magnetic fields generated by firing neurons may influence neural coordination and dynamics, we will develop a quantum-inspired computational model in which spin-sensitive MT dynamics encode memory and respond to weak magnetic fields. We will test this model using superconducting quantum circuits. This high-risk project tests a novel quantum brain mechanism poised to transform neuroscience, enable therapies, and advance AI.

An interdisciplinary team with expertise in physics, chemistry, cancer biology, oncology, and immunology has come together to overcome the physical, chemical, and biological barriers in pancreatic cancer that hindered immunotherapy.

Immune checkpoint inhibitors (ICIs) showed promise in activating adaptive T-cells to eliminate cancers, but have limited success in immune-cold tumors like pancreatic cancer. Effective ICI treatment requires a) positive tumour PD-L1 expression for PD-1/PD-L1 inhibitors to work, b) presence of tumor antigens to activate T-cells, and c) infiltration of activated T-cells. Although many pancreatic tumors express PD-L1, tumour antigen scarcity and cancer associated fibroblasts (CAFs) surrounding the tumor prevented T-cell penetration.

This proposal integrates expertise across multiple disciplines to increase tumour antigens and modify CAFs to overcome these barriers.

Objectives:

1. Evaluate irradiated autologous pancreatic cancer vaccines to generate tumour antigens outside the protected tumor microenvironment.

2. Modify CAFs using targeted radiation or fibroblast activation protein inhibitors (FAPI) to enhance T-cell access.

3. Combine cancer vaccine and CAF-modifier with ICI and optimize timing for maximal immune effect.

Approach:

Physicists and chemists will optimize radiation and FAPI delivery; cancer biologists will provide a clinically relevant (KPC) pancreatic cancer mouse model; immunologists will assess immune activation; oncologists will guide therapeutic integration.

For Objective 1, irradiated KPC cancer cells will be injected into non-tumor-bearing immunocompetent C57BL/6 mice, and T-cell populations measured. For Objectives 2 and 3, orthotopic KPC tumors in C57BL/6 mice will be treated using image-guided radiation or FAPI. In objective 2, CAF changes will be assessed via immunohistochemistry and FAPI-PET imaging, and activated T-cells in tumours quantified. Objective 3 will optimize and evaluate the combined impact of vaccine and CAF-modifier on ICI effectiveness.

Impact:

Previous studies combined ICIs with vaccines, radiation, or CAF-targeting agents individually. By employing and optimizing two strategies, we aim to overcome the two distinct barriers to ICI in pancreatic cancer. This high-risk, interdisciplinary approach could unlock effective immunotherapy for immune-cold tumors and transform pancreatic cancer treatment.

Canada’s agri-food sector faces a pressing dual challenge: producing more food while reducing carbon emissions - amid growing water scarcity and persistent labour shortages. Global market demands are pushing producers to optimize yields, while climate change calls for urgent adoption of sustainable practices that reduce water and energy use. At the same time, domestic labour gaps and rising criticism of Canada’s reliance on temporary migrant labour likened to “modern slavery” underscore the need for bold, unconventional strategies.

This high-risk, high-reward research initiative invites an unusually diverse team: scientists and engineers working alongside social scientists, educators, agri-food industry leaders, and community experts. Such a combination is rare and is necessary to drive sustainable innovation that does not just focus on technology, but also on people.

By merging experts in chemistry, materals science, education, and social sciences, we aim to accelerate both the development and adoption of advanced sensing technologies in agriculture to improve efficiency and reduce emissions. Furthermore, our goal is to ensure that these technologies improve working conditions of the labour force, provide them with opportunities for skill acquisition, and are viewed as beneficial (and thus acceptable) by growers. Our interdisciplinary team will: (1) Assess current sensing technologies, evaluating their impact on working conditions and their potential to boost productivity, cut emissions, and ensure food safety; (2) Develop next-generation sensors, designed with worker health, safety, and dignity in mind—alongside improved crop system performance; (3) Identify and overcome adoption barriers, from economic constraints to cultural resistance, ensuring these technologies can be effectively scaled across diverse farming operations; and (4) Design global training frameworks, expanding agricultural skills development in Canada and establishing new pathways to permanent residency for temporary workers.

This research lays the groundwork for a climate-resilient, technologically advanced, and socially equitable Canadian agri-food sector. Its success would place Canada at the forefront of sustainable agriculture worldwide, proving that innovation can flourish at the intersection of science, policy, and human rights.

This research addresses the cognitive and usability challenges of current assistive devices for gait, particularly 4-wheeled walkers. Despite widespread use, these devices often require significant cognitive effort and manual control, leading to reduced activity and participation, and increased abandonment. The objective is to develop a novel, user-centered smart walker that enhances mobility, safety, and social engagement by sensing, learning, and adapting to the user and their environment. From the outside, the smart walker will resemble a standard 4-wheeled walker in appearance, weight, and basic operation. However, it will be significantly smarter and more helpful, offering adaptive support and intuitive control without altering the familiar form factor.

The project uses a co-design methodology where Phase 1 involves focus groups with users and clinicians to discuss their experience with walking and gait rehab. Phase 2 involves development and initial user testing of a smart walker prototype with patent-pending electro-adhesive brakes and embedded sensors powered by an energy-efficient system, and machine learning algorithms that interpret user intent and environmental context. Phase 3 involves further refinement and testing through lab and real-world assessments, guided by user feedback and gait kinematic data.

The innovation lies in combining a breakthrough in hardware engineering (i.e., our novel electro-adhesive brakes) with advanced AI algorithms to dynamically adjust behavior in a real-time, user-responsive assistive device. This goes beyond traditional engineering-AI collaborations by incorporating mobility science and lived user experience through a co-design framework. The project also integrates qualitative methods and gait biomechanics to inform algorithm development, an approach rarely seen in engineering-led assistive device research. This convergence of technical design, human-centered design, and real-world usability creates a unique interdisciplinary space that meets and exceeds the expectations of the program.

The expected significance includes advancing assistive technology design, reducing fall risk, and improving quality of life for individuals who rely on mobility aids. Potential applications include clinical rehabilitation, community mobility, and fall prevention strategies. The project will contribute to knowledge translation across disciplines and train highly qualified personnel in cutting-edge interdisciplinary research.

Objectives

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest cancers, with survival rates below 10%, projected to be the 2nd leading cause of cancer death. Its fibrotic and “immune-cold” tumor microenvironment (TME) blocks both drugs and immune cells, leaving most patients unresponsive to immunotherapy. The protein PIN1 acts as a “master switch” that fuels tumor growth, fibrosis, and immune evasion. Our recent work published on Cell shows that blocking PIN1 can reprogram the TME and make PDAC eradicable by immunotherapy in preclinical models for the first time. However, effective Pin1-selective inhibitors are still lacking.

Hypothesis: AI-driven drug discovery can generate potent, selective PIN1 inhibitors that overcome immune resistance in PDAC by converting immune-cold tumors into immune-responsive ones.

Research Approach

This project will unite artificial intelligence (AI) drug discovery with cancer biology in a tightly integrated co-drug discovery loop.

• Aim 1 – AI-Driven Discovery: We will develop a next-generation AI model trained on the world’s largest PIN1-targeted compound dataset (~400,000 molecules generated in-house). This bias-aware model will uncover potent, chemically diverse and drug-like Pin1 inhibitors, including novel structures missed by conventional screening.

• Aim 2 – Integrated AI–Biology Validation Loop: Top AI-predicted compounds will be tested in PDAC models to assess their ability to bind PIN1, disrupt its function, and “heat up” the TME to enable immune attack. Experimental results will be fed directly back into the AI system, refining predictions and accelerating discovery toward clinically relevant therapies.

Novelty, High Risk, and Significance

This project is high-risk because it targets a historically “undruggable” protein using an novel AI-powered, bias-correcting strategy coupled to real-time experimental validation. It is high-reward because success could produce the first therapy to make PDAC eradicable by immunotherapy, transforming care for one of the most treatment-resistant cancers. The work brings together leaders in AI algorithms, AI-enabled drug discovery, and translational cancer biology in a real-time collaborative loop where computational insights and biological results continuously drive one another. This synergy enables the pursuit of a bold idea that could fail, but if successful, could redefine how resistant cancers are treated and establish a generalizable platform for drug discovery.

Following the contamination of municipal water in Walkerton in 2000, jurisdictions across Canada adopted a multi-barrier approach to protecting drinking water from “source to tap.” Policy and research has focused on water quality at the source and distribution systems. Water quality at the tap is difficult to study and address, particularly in rentals where the building owner is not the primary water consumer. Multi-unit housing constructed prior to 1960 is more likely to have plumbing with lead. Newly constructed and high-rise housing can also have water quality problems. In January 2024 over 200 tenants were evacuated from a newly built student apartment building in Hamilton Ontario because of poor water quality. There are overlooked social dimensions to addressing water quality problems. This project takes an interdisciplinary approach to studying water quality that brings together expertise from engineering, housing, and lived experience to address both the technical and social issues related to water quality in multi-unit rental buildings.

Phase 1 explores water quality through testing and interviews with tenants, building owner/managers. Phase 2 implements technologies to monitor and improve water quality at the point of use (e.g. sensors, treatment). Phase 3 employs triple bottom line analysis and fosters collaboration in creating a new water safety framework for rental buildings.

The key outcomes of this research will be (1) new information about water quality and tenant experiences in rental buildings, (2) novel technologies that will monitor and improve drinking water quality in rental buildings, (3) a dynamic data-based framework for improving water safety in these buildings, and (4) consensus on actions needed to safeguard water quality. In addition to exploring a complex, underexplored topic using mixed research methods, the research is high risk because implementation of new technologies and establishing consensus between landlords and tenants might fail. It is high reward because of the potential substantive improvements to drinking water and tenant wellness. The outcomes of this research will be applicable to other contexts and technologies will be appropriate for commercialization. The research is feasible because of the expertise of the applicants, which spans drinking water engineering, housing, and urban planning as well as the participation of tenant and industry partners.