Monocytes affect bone mineral density in pre- and postmenopausal women through ribonucleoprotein complex biogenesis by integrative bioinformatics analysis

Osteoporosis is one of the most common metabolic bone disease among pre- and postmenopausal women. As the precursors of osteoclast cells, circulating monocytes play important role in bone destruction and remodeling. The aim of study is to identify potential key genes and pathways correlated with the pathogenesis of osteoporosis. Then we construct novel estimation model closely linked to the bone mineral density (BMD) with key genes. Weighted gene co-expression network analysis (WGCNA) were conducted by collecting gene data set with 80 samples from gene expression omnibus (GEO) database. Besides, hub genes were identified by series of bioinformatics and machine learning algorithms containing protein-protein interaction (PPI) network, receiver operating characteristic curve and Pearson correlation. The direction of correlation coefficient were performed to screen for gene signatures with high BMD and low BMD. A novel BMD score system was put forward based on gene set variation analysis and logistic regression, which was validated by independent data sets. We identified six modules correlated with BMD. Finally 100 genes were identified as the high bone mineral density signatures while 130 genes were identified as low BMD signatures. Besides, we identified the significant pathway in monocytes: ribonucleoprotein complex biogenesis. What's more, our score validated it successfully.