Identification of hub mRNA, miRNAs and LncRNAs of uveal melanoma with weighted gene correlation network analysis

Abstract

To facilitate the prompt diagnosis of recurrence and treatment of uveal melanoma (UVM), the potential diagnostic markers of UVM need to be identified. The Cancer Genome Atlas (TCGA) database offers not only RNA sequence data on UVM but also the clinical characteristics of patients. The weighted gene co-expression network analysis (WGCNA) was conducted for the construction of the co-expression modules, based on how modules correlate with clinic characteristics were explored. The co-expression modules comprised of 6023 mRNA, 632 miRNA and 3042 LncRNA in 80 samples were identified, which closely associated with several clinical characteristics. A negative correlation with blue and brown modules in mRNA was revealed by overall survival (OS) time, disease free survival (DSS) time and progression free interval (PFI) time, while yellow, green and grey modules demonstrated a positive correlation. Modules turquoise in miRNA and module blue in LncRNA were most involved in survival time. The hub gene in mRNA-miRNA-LncRNA, PLCD1, LZTFL1, LCP2, GEM, NUSAP1, miR-937 and ENSG00000249846, is essential for the development of UVM. By providing a framework of co-expression gene modules for UVM, this paper found out about several diagnostic markers that could assist in the identification of recurring UVM and the treatment of it.