Balancing Complexity and Utility

Robust Cell-Type Specific Associations with Alzheimer’s Disease Within Excitatory and Inhibitory Neuronal Subclasses (ROS/MAP)

Cortical neuron loss is a pathological hallmark of late-onset Alzheimer’s disease (AD). However, despite recent single-cell transcriptomic analyses of human AD brain tissue, it remains unclear which neuronal subtypes are most vulnerable to degeneration. Here we applied cell-type proportion estimation methods to multiple RNA sequencing (RNAseq) datasets derived from human postmortem brain to identify the neuronal subtypes most strongly associated with AD.

Testing a polygenic score for microglial activation in three separate cohort studies (ADNI, CLSA, UK Biobank)

Robust associations between activated microglia and Alzheimer's disease neuropathology has been previously shown in ROS/MAP. This study aims to use the genetic architecture of microglial activation to augment predictive models of Alzheimer's disease in a statistically robust fashion.

Cell-type specific polygenic risk scores for Alzheimer's disease (ROS/MAP)

In collaboration with Dr. Young-Pearse, we calculated polygenic risk scores using genes expressed specifically by induced neuronal and glial cell. This novel cell-type specific approach can disentangle the contributions of cell type proportions to A;zheimer's disease.

Identifying psychosocial and ecological determinants of enthusiasm in youth using machine learning

Youth wellbeing is a public health priority. Enthusiasm is a known component of wellbeing and positive affectivity. Using the 2019 OSDUHS survey responses, an Ontario-wide student survey that includes many factors (demographics, drug and alcohol use, physical and mental wellbeing measures, and many more), we applied XGBoost and Shapley Additive Explanations (SHAP) to determine which factors contribute to greater adolescent enthusiasm, and therefore wellbeing. We have found that higher self-rated physical health, family relationships, and school experience are the top contributors to greater enthusiasm in youth.

Study of nicotinic receptor SNPs relationship to cognition and neuropathology in aging (ROS/MAP)

Currently utilising the ROS/MAP database to study relationships between polymorphisms in the a5 nicotinic subunit gene CHRNA5, and aging and Alzheimer's disease phenotypes.
We found a possible role of CHRNA5 in regulating amyloid processing and toxicity in a cell-type specific manner.

Association of whole-person polygenic component scores with Alzheimer's disease pathology (ROS/MAP, UK Biobank)

Late-Onset Alzheimer’s Disease (LOAD) is a heterogeneous disorder with complex etiology and high, largely unexplained, heritability. We used Pan-UK Biobank Consortium GWAS summary statistics to calculate polygenic scores (PRS) for 2,312 heritable traits in a sample of 2,044 elderly with clinical and post-mortem autopsy data. Weighted gene co-expression network analysis (WGCNA) was used to identify discrete correlated clusters of PRS. Eigen-PRS (ePRS) were calculated as the first principal component of PRS within each cluster. These ePRSs were then used to identify the module assocations with aging phenotypes within the ROS/MAP study cohort.

The Peripheral Monocyte Transcriptome is Sensitive to Microglial States in the Aging Brain (ROS/MAP)

tbc

The Caribbean-Hispanic Alzheimer’s Brain Transcriptome Reveals Ancestry-Specific Disease Mechanisms

Ethnicity impacts Alzheimer’s disease risk, especially among Caribbean-Hispanics. We report the first RNA-sequencing analysis of brain tissue from 45 Alzheimer’s disease and control Caribbean-Hispanics. Data were compared with two independent samples of non-Hispanic Caucasians (total n=729). By identifying and characterizing those genes with ancestry- and region-specific expression patterns in Alzheimer’s disease, we reveal molecular insights that may help explain epidemiological disparities in this understudied aging population.

Modelling risk of Cardiovascular diseases in bipolar disorder patients (UK Biobank)

Bipolar disorder is associated with a higher incidence of cardiovascular comorbidities, especially among females, but the mechanisms governing this sex-specific pattern of risk are unknown. This study aims to identify the genetic and neural systems that are associated with cardiovascular comorbidities in bipolar disorder among females and males, and to identify the degree to which these mechanisms are shared among other psychiatric diagnoses and across ranges of subsyndromal mood-related symptoms in the general population.

Identifying Molecular and Environmental Factors That Correlate with the Appearance of Co-morbidities in Alzheimer’s Disease (ROS/MAP, CLSA)

Mortality among patients with Alzheimer's disease (AD) is largely attributable to illnesses that arise alongside AD. Even so, the mechanisms driving these co-morbidities, and their impacts on the trajectory of cognitive decline, remain unknown. Using complex statistical modeling and machine learning, we are mapping the trajectories of of multi-morbidity as they relate to the trajectory of cognitive decline to understand the genetic and environmental factors promoting or protect against cognitive decline, and to improve AD risk prediction.

Multi-omics integration via similarity network fusion to detect subtypes of aging (ROS/MAP)

Aging is heterogeneous process among different individuals and often associated with cognitive functions. Through a data integration method – Similarity Network Fusion (SNF) patient similarity networks derived from post-mortem multi-omics data will be used to subtype individuals’ aging status and cognitive functions. This study is aiming to understand aging mechanism and related disease from a patient network level.

Modelling trajectories of psychosis spectrum symptoms in treatment-seeking youth in a naturalistic hospital setting (TAY)

tbc

Modelling major depression treatment in a structure care pathway at CAMH

tbc

The Cognitive Dysfunction in the Addictions (CDiA) Study

This study involves a multidisciplinary examination of the role of executive function deficits in a variety of outcomes among clients receiving treatment for alcohol and substance use disorder.