Description
In this study, we analyzed transcriptome gene expression microarray, epigenomic miRNA microarray and methylome sequencing data simultaneously in PBMs from 5 high hip BMD subjects and 5 low hip BMD subjects. Through integrating the transcriptomic and epigenomic data, firstly we identified BMD-related genetic factors, including 9 protein coding genes and 2 miRNAs, of which 3 genes (FAM50A, ZNF473 and TMEM55B) and one miRNA (hsa-mir-4291) showed the consistent association evidence from both gene expression and methylation data, and 3 genes (TMEM55B, RNF40 and ALDOA) were confirmed in the meta-analysis of 7 GWAS samples and GEnetic Factors for OSteoporosis consortium (GEFOS-2) GWAS results. Secondly in network analysis we identified an interaction network module with 12 genes and 11 miRNAs including AKT1, STAT3, STAT5A, FLT3, hsa-mir-141 and hsa-mir-34a which have been associated with BMD in previous studies. This module revealed the crosstalk among miRNAs, mRNAs and DNA methylation and showed four potential regulatory patterns of gene expression to influence the BMD status, including regulation by gene methylation, by miRNA and its methylation, by transcription factors and co-regulation by miRNA and gene methylation. In conclusion, the integration of multiple layers of omics can yield more in-depth results than analysis of individual omics data respectively. Integrative analysis from transcriptomics and epigenomic data improves our ability to identify causal genetic factors, and more importantly uncover functional regulation pattern of multi-omics for osteoporosis etiology.