Identification of Co-Expression Modules and Genes Associated With Tumor Progression in Oral Squamous Cell Carcinoma.

Abstract

Oral squamous cell carcinoma (OSCC) is a common head-and-neck cancer with a deficiency of early diagnosis and poor prognosis. To identify potential diagnostic and prognostic markers of OSCC, we firstly used weighted gene co-expression network analysis (WGCNA) to build a co-expression module from GSE42743. Next, we performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses on specified units from selected modules utilizing Database for Annotation, Visualization, and Integrated Discovery (DAVID). Additionally, we identified and validate hub genes of these specified modules from multiple datasets like GEPIA and TCGA. In total 16 co-expression modules were built by 17,238 genes of 74 tumor samples utilizing WGCNA. Through pathway and functional enrichment analysis, the turquoise module was most firmly relevant to the cell cycle, oocyte meiosis, and p53 signaling pathway. Hub genes VRK1, NUP37, HMMR, SPC25, and RUVBL1 were identified to be related to oral cancer at both molecular level and clinical levels. The expressions of these genes differed in tumor tissues and normal tissues. Meanwhile, patients with high hub gene expression had a poor prognosis clinically. To conclude, five hub genes were identified to be relevant to oral cancer from the molecular level and the clinical level. Therefore, the detection of these genes was of great significance. They can be regarded as diagnostic and prognostic biomarkers for oral cancer. Also, they could shed light on the improvement of patients' overall survival and prognosis, which needs further analysis in the future.