Biological Functions And Prognostic Value Of RNA-Binding Proteins In Bladder Cancer

Objective:Bladder cancer(BLCA)is a malignant tumor with high mor-bidity and high mortality.The steadily-rising incidence and prevalence make bladder cancer one of the most common urogenital cancers in the word.The abnormal expression of RNA binding protein(RNA-binding Proteins,RBPs)has been reported in many kinds of malignant tumors and is closely related to the occurrence and development of tumors.However,the role of RNA-binding protein in bladder is still unclear,so it is very meaningful to explore the potential biological function of RNA-binding protein in bladder cancer and establish a clinical predictive model.Methods:In this study,the RNA sequencing data and clinical data of BLCA were downloaded from the tumor genome map(TCGA)database through bioinformatics analysis,and the differentially expressed RNA binding protein(DERBPs)in normal and cancer tissues was screened by R software.In order to identify the related pathways and annotate the function of DERBPs,the R software was used for GO&KEGG bioaccumulation analysis to analyze the main biological processes involved and visualized.After that,the protein-protein interaction(PPI)was analyzed by using the interactive gene retrieval tool(STRING)database,and the key modules were screened and visualized by Cytoscape3.7.2 software and its MCODE plug-in.After univariate and multivariate Cox regression analysis of DERBPs gene,the RBPs,related to survival was identified and the prognostic model was established.In order to verify the validity of the model,the survival data of all patients were randomly divided into training group(Train)and verification group(Test,)at 1:1 to calculate the risk scores of the two groups.The validity of the model was verified by survival analysis,ROC curve,risk curve and independent prognosis analysis.Then the line diagram is drawn based on the prediction model.Finally,the Human Protein Atlas(HPA)database was used to verify the effectiveness of the model again at the gene transcription and translation levels.Results:A total of 385 DERBPs genes were screened,including 218 up-regulated genes and 167 down-regulated genes.GO&KEGG bioenrichment analysis suggested that DERBPs was mainly involved in nc RNA metabolic process,nc RNA processing,ribonucleoprotein complex biogenesis,Regulation of translation,RNA splicing,t RNA metabolic process,Ribosome biogenesis,RNA catabolic process,m RNA catabolic process,cytoplasmic ribosome,Regulation of m RNA metabolic process,ribonucleoprotein particles,RNA catalytic activity and so on..When the confidence score is set to 0.7 in the STRING database,a protein interaction network with 311 nodes and 2020 edges is obtained.After that,the key sub-networks of three PPI networks are screened by using the MCODE plug-in in Cytoscape3.7.2 software,and the visualization and function enrichment analysis are carried out.After univariate and multivariate Cox regression analysis of DERBPs gene,the best prediction model was determined,which was composed of eight DERBPs,such as CLK2,HNRNPH3,CTU1,OAS1,MRPL38,CTIF,DARS2 and XPOT.Further analysis according to the model shows that there is a significant correlation between patients’ risk score and their prognosis.The analysis shows that the area under the ROC curve of the 1-year,3-year and 5-year risk value of the 8-RBPs prediction model is 0.742,0.713 and 0.776.After that,a line chart based on 8-RBPs prognostic risk score model is established,which can well predict the 1-,3-and 5-year overall survival rate of BLCA patients.In order to verify the validity of the model,the internal verification was carried out in the TCGA cohort.The AUC of the validating group’s 1-year,3-year and 5-year Va R was 0.720,0.639 and 0.653 respectively,which also verified the predictive value of the model to BLCA.Finally,the Human Protein Atlas(HPA)database was used to verify the effectiveness of the 8-RBPs model at the transcription and translation levels.Conclusions:Based on bioinformatics methods,through BLCA analysis of TCGA database,the best prediction models composed of CLK2,HNRNPH3,CTU1,OAS1,MRPL38,CTIF,DARS2 and XPOT were obtained.The reliability of 8-RBPs gene prognosis model for predicting BLCA was verified by multiple indicators,and a line map was constructed based on 8-RBPs gene.This study will help to improve the understanding of the pathogenesis and prognosis of BLCA and provide reference for further discussion of the diagnosis and treatment of BLCA.