Our research identifies how genes influence disease risk by integrating lots of big datasets. We use health information on thousands and sometimes millions of people, coupled with data on genetic and epigenetic variation across the whole genome, to advance precision health. These big datasets give us a more complete picture of a person’s health and environment, which helps us to zoom in on the particular elements that are contributing to disease. Once we have identified these particular elements, we can move towards specific personalized prevention strategies and therapies.
Genetics of Brain-Related Disorders
Brain-related disorders such as depression and Alzheimer’s disease account for more years of life lost to disability and death than either cancer or cardiovascular disease. Three quarters of psychiatric disorders manifest in childhood and adolescence, and brain-related disorders affect not only the person with the disorder, but also their family and support systems as well. Our research is directed towards reducing the suffering associated with psychiatric and neurological disorders by understanding genetic influences on these traits.
We have studied genetic risk factors for loneliness, depression, and schizophrenia in large consortia, including the PsycheMERGE consortium. PsycheMERGE is a growing network of clinical sites with electronic health records linked to genomic data, which we leverage for psychiatric genetics research.
We also have ongoing projects on the role of genetics in traumatic brain injury recovery. Concussions are sustained by one in 150 Canadians each year. In the short term, people with concussions may struggle to resume pre-injury activities months after the injury. In the long term, concussions have been linked to dementia and Alzheimer’s disease. We are looking for genetic risk profiles that help us identify people at risk of poor recovery soon after their injury, so that we can provide treatment before these bad outcomes arise.
Harnessing “Big Data” to Advance Precision Health
Data collected in routine clinical care are a huge source of potentially valuable data. These data include electronic health records (EHRs) and other health administrative records, such as information on prescriptions filled and laboratory test results. The Dennis lab analyzes these data using advanced computational approaches in order to advance precision health. We apply machine learning techniques to clinical data so that we can identify data-driven patient groupings. These groupings often reveal disease patterns that we didn’t know existed. In this way, we’re moving beyond a one-size-fits all disease classification system, towards one that’s more tailored to each person’s individual data.
Clinical data linked to other large genomic and environmental datasets are also leading to advances in precision health. We are applying statistical genetics methods for the prediction and early detection of brain-related disorders in the US-based PsycheMERGE consortium. In Canada, we are moving towards similar approaches. BC Children’s Hospital has prioritized digital health research, which aims to harness data and technologies to improve health outcomes. The Dennis lab is on the forefront of these digital health innovations. We will play a key role in developing data resources combining clinical, genomic, and environmental data, to advance patient-centered, precision health in Canada.
Genomic Data Integration
New technologies are allowing us to query the human body like never before. We can now measure genetic variation across the whole genome, and relate it to gene expression and epigenetic marks in multiple cells and tissues. These measurements have revealed that the interactions between molecules in our bodies are hugely complex and dynamic, changing over time and in response to different environments.
Studying the relationships between different genomic molecules (such as genes expressed and epigenetic marks) will help us understand the basis of health and disease. The Dennis lab uses computational methods that integrate different datasets to learn about disease processes. For example, collaborators studying gene regulation recently discovered that some genomic regions were more versatile than we previously thought, specifically, that some gene promoters could double as gene enhancers. We are now are exploring how mutations in these genomic regions impact human health and disease susceptibility by integrating data on gene regulation with data on human health in large biobank datasets.