Screening and identification of key biomarkers and related transcription factors in ovarian cancer by integrated bioinformatics analysis
Wenqiong Qin,Qiang Yuan,Yi Liu,Ying Zeng,Dandan Ke,Xiaoyan Dai,Yu Shuai,Jiaqi Hu,Hua Shi
DOI: https://doi.org/10.21203/rs.3.rs-20327/v1
2020-04-03
Abstract:Abstract Background Ovarian tumors are the most malignant tumors of all gynecological tumors, and although multiple efforts have been made to elucidate the pathogenesis, the molecular mechanisms of ovarian cancer remain unclear. Methods In this study, we used bioinformatics to identify genes involved in the carcinogenesis and progression of ovarian cancer. Three microarray datasets (GSE14407, GSE29450, and GSE54388) were downloaded from Gene Expression Omnibus (GEO) database, and differentially expressed genes (DEGs) were identified. For a more in-depth understanding of the DEGs, functional and pathway enrichment analyses were performed and a protein-protein interaction (PPI) network was constructed. The associated transcriptional factor (TFs) regulation network of the DEGs was also constructed. Kaplan Meier-plotter, Gene Expression Profiling Interactive Analysis (GEPIA), the Human Protein Atlas (HPA) database and the Oncomine database were implemented to validated hub genes. Results A total of 514 DEGs were detected after the analysis of the three gene expression profiles, including 171 upregulated and 343 downregulated genes. Nine hub genes ( CCNB1, CDK1, BUB1, CDC20, CCNA2, BUB1B, AURKA, RRM2, TTK) were obtained from the PPI network. Survival analysis showed that high expression levels of seven hub genes ( CCNB1, BUB1, BUB1B, CCNA2, AURKA, CDK1, and RRM2) were associated with worse overall survival (OS). All of seven hub genes were discovered highly expressed in ovarian cancer samples compared to normal ovary samples in GEPIA. Immunostaining results from the HPA database suggested that the expressions of CCNB1, CCNA2, AURKA, and CDK1 proteins were increased in ovarian cancer tissues, and Oncomine analysis indicated that the expression patterns of BUB1B, CCNA2, AURKA, CCNB1, CDK1, and BUB1 have associated with patient clinicopathological information. From the gene-transcriptional factor network, key transcriptional factors, such as POLR2A, ZBTB11, KLF9, and ELF1, were identified with close interactions with these hub genes. Conclusion We identified six significant DEGs with poor prognosis in ovarian cancer, which could be of potential biomarkers for ovarian cancer patients.