Research on a Weighted Gene Co-expression Network Analysis method for mining pathogenic genes in thyroid cancer.

Abstract

Thyroid cancer (TC) is one of the most common thyroid malignancies occurring worldwide, and accounts for about 1% of all the malignant tumors. It is one of the fastest growing tumor and can occur at any age, but it is more common in women. It is important to find the pathogenesis and treatment targets of TC. In this pursuit, the present study was envisaged to investigate the effective carcinogenic biological macromolecules, so as to provide a better understanding of the occurrence and development of TC. The clinical and gene expression data were collected from The Cancer Genome Atlas (TCGA). We clustered mRNA and long non-coding RNA (lncRNA) into different modules by Weighted Gene Co-expression Network Analysis (WGCNA), and calculated the correlation coefficient between the genes and clinical phenotypes. Using WGCNA, we identified the module with the highest correlation coefficient. Subsequently, by using the differential genes expression analysis to screen the differential micro-RNA (miRNA), the univariate Cox proportional hazard regression was employed to screen the hub genes related to overall survival (OS), with P < 0.05 as the statistical significance threshold. Finally, we designed a hub competitive endogenous RNA(ceRNA) network of disease-associated lncRNAs, miRNAs, and mRNAs. From the results of enrichment analysis, the association of these genes could be related to the occurrence and development of TC, and these hub RNAs can be valuable prognostic markers and therapeutic targets in TC.