Integrated Analysis of the Functions and Prognostic Values of RNA Binding Proteins and Candidate Drugs in Renal Papillary Cell Carcinoma

Abstract

Background: Roles of RNA binding proteins in renal papillary cell carcinoma (KIRP) remain undiscovered. We thus conducted a series of bioinformatics analyses to elucidate the associations between RBPs and prognosis of renal papillary cell carcinoma.Methods: RNA sequencing data and clinical of KIRP were downloaded from the TCGA database. The differentially expressed RBP coding genes (DEGs) were sorted out by R software and Gene Ontology (GO) function and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were then performed to evaluate the functional pathways. Protein-protein interaction (PPI) network of DEGs was formed through the Search Tool for the Retrieval of Interacting Genes (STRING) database and visualized by Cytoscape. Subsequently, a prognostic model was constructed by the uses of univariate Cox regression analysis, random survival forest analysis and multivariate Cox analysis. Validations of Receiver Operating Characteristic (ROC) analysis, KM analysis of overall survival and nomogram were performed as follow. Furthermore, CMap database was used to predict potential drugs.Results: A prognostic OS-predictive model based on six RBPs (SNRPN, RRS1, INTS8, RBPMS2, IGF2BP3 and PIH1D2) was constructed. STOCK1N-28457, pyrimethamine and trapidil were determined as potential drugs according to the CMap database.Conclusion: This study constructed a prognosis-related six-RBP signature and made a prediction of three small molecular drugs, which provided a novel insight into KIRP and assisted the development of individualized therapeutic strategies.