Grant: Master's Award: Canada Graduate Scholarships

microRNA (miRNAs) are small non-coding RNA molecules that bind to mRNAs to negatively regulate gene expression. miRNAs play important roles in nearly all physiological processes. Consequently, dysregulated miRNA expression is linked to the development of various diseases such as dementia and cancer. Compared to mRNAs, miRNAs are more accessible (i.e. urine, blood) and stable, while also being specific for disease-stage and cell-type. These properties make miRNAs excellent biomarker candidates. However, before identifying a miRNA as a biomarker for a condition of interest, we must address two major issues in the study of miRNA biology. There is a lack of consensus on the number and identity of many miRNAs. Sequence databases used to align and quantify miRNAs are wrought with inconsistencies, low-confidence entries, and false-positives. Databases like miRBase list over 2656 human miRNAs, but only 25% of them are considered “high confidence.” To successfully study miRNAs as biomarkers, we must first resolve which miRNAs are real and where they are expressed within the body. Then, another problem arises when trying to select only a few markers that are biologically relevant for a condition of interest, from the growing list of discovered miRNAs? To address these two challenges, we propose two projects. In project 1, we aim to validate miRNA sequences and characterize their tissue- and disease- specific expression through a comprehensive miRNA atlas. In project 2, we aim to provide a methodology for discerning biologically relevant miRNAs from large omics datasets using machine learning techniques. Together, these projects provide essential tools and resources for the progression of miRNA research and their biomarker applications.