Construction of osteoporosis diagnosis model based on bioinformatics analysis of autophagy-related genes.

Osteoporosis (OP) is a major health issue that poses a substantial challenge to public health worldwide. In recent decades, knowledge about the etiological mechanisms of osteoporosis has emphasized that bone cell homeostasis is tightly regulated by autophagy. However, little is currently known about autophagy-related genes (ATGs) that influence osteoporosis, which compromises our deep understanding of the pathogenesis of osteoporosis and our ability to develop targeted treatment strategies. Here, we screen differentially expressed autophagy-related genes (DE-ATGs) in osteoporosis using GSE56815 dataset and human autophagy database and explore the role of autophagy-related biomarkers in the occurrence and development of OP to provide new ideas for the diagnosis and treatment of OP. The original dataset was downloaded from the gene expression omnibus database and further integrated and analyzed. The differential genes of OP were screened using R software, and the differentially expressed genes of OP were intersected with ATGs obtained from the human autophagy gene pool to obtain DE-ATGs. The differentially expressed autophagy genes were enriched using gene ontology, Kyoto Encyclopedia of Genes and Genomes, Metascape, DisGeNet, and gene set enrichment analysis. Core DE-ATGs were screened using Least Absolute Shrinkage and Selection Operator and cross-validation, and an receiver operating characteristic curve was constructed to explore the diagnostic value of key genes for OP. According to receiver operating characteristic curve analysis, ras-related C3 botulinum toxin substrate 1, epidermal growth factor receptor, and cathepsin D are key genes with high diagnostic value related to autophagy, which can provide a new method for the diagnosis and treatment of OP.