Informing Osteoporosis GWAS Using Networks

点击翻译按钮获取中文摘要

Project Summary: Osteoporosis is a complex disease characterized by low bone mineral density (BMD), bone fragility, and an increased risk of fracture. Genome-wide association studies (GWASs) for BMD have identified over 1100 associations. This “treasure trove” of novel genetic information has the potential to revolutionize our understanding of bone biology and the treatment of bone diseases; however, few of the causal genes underlying associations have been identified. The long-term goal of my lab is to reduce this knowledge gap using innovative analytical and experimental strategies. A number of approaches exist to identify genes responsible for GWAS associations. However, most rely on population-based “-omics” data, which are scarce for human bone, to connect variants to molecular alterations. Furthermore, most approaches do not provide information on how causal genes impact “systems-level” function. To address these limitations, we recently used co-expression networks generated from mouse bone transcriptomic datasets to inform BMD GWAS. The idea is simple – genes that play a central role in the regulation of a complex trait are often functionally-related and functionally-related genes are often co-expressed. We demonstrated that by identifying modules of co-expressed genes in bone tissue that were enriched for genes implicated by GWAS, we were able to predict target genes and infer how they impact BMD. We have shown experimentally that several genes (MARK3, PPP6R3, etc.) identified using this approach are true regulators of BMD. However, to date our analyses have been based only on enrichment of genes implicated by GWAS in undirected networks generated from heterogenous bulk bone transcriptomic datasets of limited cellular diversity. Here, we address these limitations using an innovative analytical approach incorporating Bayesian networks and key driver analysis (KDA) and apply this strategy to transcriptomic data from primary bone cell cultures and osteoblast subtypes defined by single cell RNA-seq (scRNA-seq). We hypothesize that our approach of integrating GWAS with directed networks representing all major bone cell types will identify novel genes with direct and central roles in regulating BMD and elucidate how such genes impact network architecture. We will test this hypothesis through three specific aims. In Aim 1, we will discover novel BMD GWAS target genes through the integration of bone and bone cell networks and GWAS data. In Aim 2, we will discover novel BMD GWAS target genes through the integration of osteoblast cell-type specific networks and GWAS data. In Aim 3, we will define the impact of novel BMD GWAS target genes on BMD and bone network homeostasis. Our innovative approach for informing GWAS will identify causal BMD genes and lead to the discovery of putative therapeutic targets for the prevention and treatment of bone fragility.