To analyze the abnormally expressed genes involved in cervical cancer occurrence and development.
Integrated bioinformatics methods were used to analyze differentially expressed (DE) RNAs, including mRNAs, microRNAs (miRNAs), and long non-coding RNAs (lncRNAs), in stage I, II, III, and IV cervical cancer patients from the TCGA database to fully reveal the dynamic changes caused by cervical cancer.
First, DE RNAs in cervical cancer tissues from stage I, II, III, and IV patients and normal cervical tissues were identified and divided into different profiles. Several DE RNA profiles were down-regulated or up-regulated in stage I, III, and IV patients. GO and KEGG analysis of DE mRNA profile 1, 2, 4, 5, 6 and 22 which were significantly down-regulated or up-regulated showed that DE mRNAs are involved in cell division, DNA replication, cell adhesion, the positive and negative regulation of RNA polymerase ll promoter transcription. Besides, DE RNA profiles with significant differences in patient stages were analyzed to perform a competing endogenous RNA (ceRNA) regulatory network of lncRNA, miRNA, and mRNA. The protein-protein interaction (PPI) network of DE mRNAs in the ceRNA regulatory network was also constructed. The network had nine central genes (up-regulated genes: CDKN2A, GSK3B, BIRC5, CYCS, MAD2L1; down-regulated genes: PTEN, FOXO3, CCND2, TGFBR2). Survival analysis found that 5 lncRNAs, 9 mRNAs, and 4 miRNAs can be used as prognostic indicators of cervical cancer. Finally, combined with cluster analysis results, we further screened 2 DE RNAs (AMZ2P1 and HDAC5) using clinical samples, suggesting that AMZ2P1, and HDAC5 may act as diagnostic biomarkers for the development of cervical cancer.
This research provides new effective targets and reliable biological markers for the diagnosis and prognosis of cervical cancer.