Grant: Discovery Grants Program - Individual

Recent technical advances for measuring mitochondrial DNA variation have resulted in the comprehensive assessment of mtDNA quantity and sequence changes (heteroplasmy). However, the field of mt variant calling remains imperfect and lags behind nuclear variant calling due to the requirement for increased read depth and unique computational approaches. Our long-term objective is to understand the cellular interactions that define mitochondrial-nuclear crosstalk as well as develop and benchmark novel statistical methods for the comprehensive assessment of mtDNA and its relationship with the nuclear genome. My research program will develop a suite of data analysis tools to enable comprehensive evaluation of mitochondrial DNA variation across species and tissue types. In the short term, our work will focus on tool development for integration, imputation, and functional characterization of mtDNA variation with the cell nucleus. This will include statistical models for the integration of mtDNA quantity and mitochondrial variant calls with other multiomics datasets including nuclear transcriptomic, and epigenomic data. It will also include the use of novel machine learning algorithms to impute mtDNA quantity and mtDNA variant thresholds from nuclear multiomic signatures as well as cellular models to introduce targeted mtDNA mutations, manipulate heteroplasmic fraction, and directly modify mtDNA-CN levels in culture to determine the influence of mtDNA variation on the cell nucleus across tissues. Through extensive coordination with the nuclear genome, the mitochondrion plays an important role in translating environmental stimuli into alterations to nuclear gene expression via the nuclear epigenome, and transcriptome. Our group and others have provided evidence that pervasive and long-lasting regulation of gene expression, DNA methylation, and metabolite levels arise from mtDNA variation, suggesting a role for externally induced nuclear remodeling via modifications to mitochondrial DNA. In the context of statistical modelling, these relationships can be exploited via the integration of multiple mitochondrial and nuclear multiomic measures as well as imputation of mtDNA metrics via nuclear DNA signatures. These new tools will accelerate all kinds of biological research both academic and industry-based that will benefit from mitochondrial DNA calling, imputation, and/or consequences of the mtDNA variation levels. They will enable increased efficiency and accuracy in mtDNA calling across various biological questions including ancestral lineage tracing, environmental impacts on mtDNA, and mitochondrial homeostasis, among others. The activities outlined in this proposal will serve as the beginning of a long-term NSERC-funded research program that will generate methods for mitochondrial calling, variant integration, and characterization, and will train high quality personnel in molecular biology, genetics, and bioinformatics.