Leveraging GWAS and the Transcriptome to Identify Potential Causal Genes in Osteoperosis

Osteoporosis is a prevalent bone disease that poses a significant health problem for millions of individuals globally. Genome-wide association studies (GWASs) have identified numerous associations that affect bone mineral density (BMD), the most reliable predictor of osteoporosis fracture. Further efforts are being made to identify the genes responsible for the effects of these associations. Most of these associations impact bone by altering gene regulation. In my dissertation work, I used innovative, unbiased approaches to prioritize previously identified genetic associations from humans. In the first chapter, I discuss how molecular "-omics" data and state-of-the-art analytical techniques are being employed to facilitate gene discovery from GWAS and provide meaning to these studies. I highlight the resources required in the bone field and novel approaches that I used in my graduate work, along with their potential for improvement in the coming years. In the second chapter, I focus on identifying potentially causal long non-coding RNAs (lncRNAs), which are understudied non-coding RNAs in the context of bone and osteoporosis. I identified 23 lncRNAs that may play a causal role in osteoporosis and are candidates for experimental follow-up studies. In the third chapter, I used long-read proteogenomics to identify potentially causal protein-coding isoforms in osteoporosis. I provide a list of potentially causal isoforms and validated TPM2 functionally in vitro. Finally, I share my final thoughts on the current state of the field and future directions for the next generation of systems geneticists who seek to provide treatment for osteoporosis. Ultimately, my dissertation contributes to our comprehension of the genetic architecture of osteoporosis-related traits and presented new approaches for following up on GWAS studies.