Identification and Analysis of Genes Underlying Bone Mineral Density by Integrating Microarray Data of Osteoporosis

Abstract

Osteoporosis is a kind of brittle bone disease, which is characterized by a reduction in bone mineral density (BMD). In recent years, a number of genes and pathophysiological mechanisms have been identified for osteoporosis. However, the genes associated with BMD remain to be explored. Toward this end, we integrated multiple osteoporosis microarray datasets to identify and systematically characterize BMD-related genes. By integrating the differentially expressed genes from three osteoporosis microarray datasets, 152 genes show differentially expressed between high and low BMD osteoporosis samples in at least two of the three datasets. Among them, 88 were up-regulated in high BMD samples and 64 were up-regulated in low BMD samples. The expression of ZFP36, JUNB and TMEM8A were increased at high BMD samples in all three datasets. Hub genes were further identified by co-expression network analysis. Functional enrichment analysis showed that the gene up-regulated in high BMD were enriched in immune-related functions, suggesting that the immune system plays an important role in osteoporosis. Our study explored BMD-related genes based on the integration of osteoporosis microarray data, providing guidance to other researchers from a new perspective.

Keywords: bone mineral density; co-expression; enrichment analysis; microarray; osteoporosis.

FIGURE 1

Differential expression analysis. (A–C) Volcano plot of DEGs in each dataset, red nodes represent upregulation in high BMD samples and blue nodes represent upregulation in low BMD samples. (D) Statistics of two types of DEGs.

FIGURE 2

Integration of BMD-related DEGs. (A) UpSetR plot of genes up-regulated in high BMD samples, the number of genes present in at least two datasets is marked (red). (B) UpSetR plot of genes up-regulated in low BMD samples, the number of genes present in at least two datasets is marked (red). (C) Boxplot of genes that expressed consistently across all three datasets.

FIGURE 3

Co-expression network and function enrichment analysis of genes that are consistently expressed in high BMD samples. (A) The gene co-expression network. Node size represents how many other genes interact with it. Node color indicates in which data sets it is up-regulated in high BMD samples. The darker the edge color, the greater the correlation coefficient. (B) GO and KEGG results of genes in network.

FIGURE 4

Co-expression network and function enrichment analysis of genes that are consistently expressed in low BMD samples. (A) The gene co-expression network. Node size represents how many other genes interact with it. Node color indicates in which datasets it is up-regulated in low BMD samples. The darker the edge color, the greater the correlation coefficient. (B) GO and KEGG results of genes in network.