Combined analysis of gene expression, miRNA expression and DNA methylation profiles of osteosarcoma

Osteosarcoma (OS) is a common primary malignancy in children and adolescents with relative high survival rate after chemotherapy. While the toxicity of chemotherapy and personalized different response to chemotherapy makes it difficult for the selection of therapeutics and improvement of diagnosis. In this study, we conducted a combined analysis of three types of microarray datasets (gene expression, microRNA (miRNA) expression and DNA methylation) from the Gene Expression Omnibus (GEO). The differential expression genes (DEGs) and miRNAs (DEMI) were screened out via the limma package and differential methylation sites (DMS) were identified by the IMA package. Enriched functions of DEGs and genes contained DMS (DEMs) were obtained through the Database for Annotation, Visualization and Integrated Discovery (DAVID). Besides, miRNA-gene regulation network was obtained based on the pairs of involved DEMIs and overlapping genes between DEMs and DEGs and visualized through Cytoscape software. A total of 583 DEGs and 1051 DMS (corresponding to 827 DEMs) were identified and 56 overlaps were obtained. As expected, most of the expression and methylation profiles of the overlaps exhibited significant negative correlation. The DEGs were mainly enriched in the biological processes related to inflammatory/immune response, cell proliferation, while DEMs were involved in the regulation of gene expression, tissue/organ development. Based on the correlation and network analysis, some novel targets were identified for OS and many known biomarkers were proved in this study, which would be helpful in its early diagnosis and treatment.