Exploring CircRNA Diagnostic Markers And Regulatory Mechanisms In Colorectal Cancer By Bioinformatic Methods

?Objective?The biological function of circRNA has only been noticed in recent years.Circ RNA has the characteristics of rich expression,continuous stability,non-invasive materials and high specificity,which is suitable as a potential diagnostic marker and assist in early diagnosis in cancer.Circ RNA also plays an important role in regulating the occurrence and development of cancer and can be used as a potential therapeutic target,providing new ideas for cancer treatment.This research aims to find colorectal cancer potential circRNA diagnostic markers,and further confirm the circRNA with the most regulatory function in this study and its circRNA-miRNA-mRNA regulatory axis,and explore its regulatory mechanisms,by digging circRNA,miRNA,and mRNA public databases of colorectal cancer,combining several target prediction databases,using bioinformatic analysis methods,based on competing endogenous RNA regulatory theory.?Methods?We searched circRNA series of colorectal cancer tissues and adjacent tissues in the GEO database,and selected GSE126094 with 10 pairs of samples and GSE142837with 3 pairs of samples as the research objects.We downloaded GSE126094 expression matrix and used SangerBox software tools to normalize the data by quantiles method.Using limma package with|log₂FC|>1.5 and adj.P<0.01 as the criteria,we searched for the differential expression circRNAs in GSE126094.GSE142837 was handled in the same way.The intersection of the two series of differential expression circRNAs were the DEcircRNAs of this study.We defined the diagnostic value of DEcircRNAs through ROC curve drawn by SPSS software.We used SangerBox software tool to download colorectal miRNA expression matrix from TCGA database.The data of 616 colorectal cancer and 11 normal samples were normalized by quantiles method.Also using limma package with|log₂FC|>1,adj.P<0.01 as the criteria,we searched for the differential expression miRNAs.We predicted the miRNAs that DEcircRNAs maight bind to in circ Bank and circ Interactom databases.The intersection of the three groups was the target DEmiRNAs.GEPIA 2 database can analyze the mRNA expression data of TCGA combined with GTEx database.A total of 367 colorectal cancer samples and 359 normal samples were selected.Also using limma method with|log₂FC|>1,adj.P<0.01 as the criteria,we searched for the differential expression mRNAs.We predicted the mRNAs that the target DEmiRNAs maight bind to in Targetscan,miRDB,and mi Tarbase databases.The intersection of the four groups was the target DEmRNAs.Cytoscape software was used to drawing DEcircRNAs,target DEmiRNAs,target DEmRNAs and their targeting relationships as circRNA-miRNA-mRNA regulatory network.We used STRING database and Cytoscape software to build Protein Protein Interaction network,and used MCODE to find the key genes with the closest interaction,and found the most critical gene with the largest Degree.We used Cytoscape software again to construct the circRNA-miRNA-mRNA regulatory network of key genes.Based on the theory of competing endogenous RNA regulatory in which mRNA and miRNA expression were negatively correlated,miRNA and circRNA expression were negatively correlated,we searched for upstream circRNA of the most critical gene and its regulatory axis.g:Profiler database was used to perform GO and KEGG enrichment analysis on target DEmRNAs to find the biological function of circRNA through the regulatory axis.?Results?There were 6 DEcircRNAs in the intersection of GSE126094 and GSE142837 in GEO database,they were:hsa?circ?0006332,hsa?circ?0065173,hsa?circ?0000518,hsa?circ?0006174,hsa?circ?0087862 and hsa?circ?0005927.All of them have relatively high diagnostic value.There were 13 target DEmiRNAs,which was the intersection of differential expression miRNAs in the TCGA database and predicted miRNAs.There were 130 target DEmRNAs,which was the intersection of differential expression mRNAs from TCGA combined with GTEx database and predicted mRNAs.Discarding 2 circRNAs and 11 mRNAs because they existed independently in the network after the intersection of targeting relationships,we used 4 DEcircRNAs,13target DEmiRNAs,119 target DEmRNAs and their targeting relationships to construct the circRNA-miRNA-mRNA regulatory network.The Protein Protein Interaction network showed that these mRNAs were closely related,with CCND1,TNRC6A,KAT2B,E2F3,E2F2,and CCNE2 as the key genes,and CCND1 as the most critical gene.We used 6 key genes and their upstream 8 miRNAs,3 circRNAs and their targeting relationships to construct the circRNA-miRNA-mRNA regulatory network of key genes.From this network,we obtained hsa-miR-892b,which was negatively correlated with CCND1,and hsa?circ?0087862,which was negatively correlated with hsa-miR-892b.Therefore,hsa?circ?0087862 was the circRNA with the most regulatory function in this study,and its regulatory axis was hsa?circ?0087862/hsa-miR-892b/CCND1.The up-regulated hsa?circ?0087862,sponging hsa-miR-892b,weakening the translational inhibition of CCND1 by hsa-miR-892b,thereby promoted the expression of CCND1.Target DEmRNAs enrichment analyzing showed that CCND1 was enriched in protein binding in molecular functional,positive regulatory of biological processes and cellular processes in biological process,anchoring junction in cellular component,cell cycle in KEGG and so on.?Conclusions?Hsa?circ?0006332,hsa?circ?0065173,hsa?circ?0000518,hsa?circ?0006174,hsa?circ?0087862 and hsa?circ?0005927 can work as potential circRNA diagnostic markers in colorectal cancer.Hsa?circ?0087862 is the colorectal cancer circRNA with the most regulatory function in this study,which promotes the occurrence and development of colorectal cancer through hsa?circ?0087862/hsa-miR-892b/CCND1regulatory axis,by exerting protein binding function,participating in positive regulatory of biological processes and cellular processes,acting on anchoring junction and undergoing cell cycle pathway.