Study On The Relationship Between Chemokines And Prognosis Of Ovarian Cancer Based On Bioinformatics Methods

Objective:We analyzed and mined the prognostic chemokines of ovarian cancer using bioinformatics methods,constructed a multi-gene prognostic-related prediction model,and explored the significance of chemokines in the prognosis of ovarian cancer,in order to provide new reference for the optimization of ovarian cancer prevention and treatment plan.Method:First,the transcriptome data and clinical data of ovarian cancer tissue samples and normal ovarian tissue samples were downloaded from the public database of cancer genome map(TCGA)and genotypic tissue expression(GTEx)for principal component analysis,and the differentially expressed genes in ovarian cancer tissue and normal ovarian tissue were screened using limma package in Bioconductor of R language,and the differential gene clustering heat map and GO enrichment analysis were drawn using pheatmap package and ClusterProfiler package.Then,the intersect function of R language was used to mine chemokines with differential expression in the differentially expressed genes,and the survfit function was used for Kaplan-Meier(K-M)survival analysis to screen for chemokines with differential expression related to prognosis.The obtained chemokines were used to construct and optimize a Cox model.With the model,ovarian cancer samples were divided into high and low risk groups by using median values,and then the ROC curve was used to verify the effectiveness of the Cox model.Finally,the CIBERSORT algorithm was used to analyze the proportion of immune cells in ovarian cancer and the proportion of different types of immune cells in high-risk and low-risk groups than those in the normal ovarian cancer group.In addition,the correlation between the immune cells and chemokines in the model was analyzed by using R language rcorr function.Result:1.The R software limma package analyzed the gene expression levels in ovarian cancer and normal ovarian tissue samples,and found that there were a total of 6813 differentially expressed genes between ovarian cancer and normal ovarian tissue samples,including 3085 up-regulated genes and 3728 down-regulated genes.2.The R language ClusterProfiler performs GO enrichment analysis on 6813 differentially expressed genes.The GO enrichment analysis includes biological processes(BP),cellular components(CC)and molecular functions(MF)..BP related content includes extracellular matrix tissue,extracellular tissue,cell adhesion,cell adhesion,etc.;CC related to collagen extracellular matrix,cell connection,etc.;MF related to extracellular matrix structure,glycosaminoglycan binding,muscle Kinesin binding and so on.3.Using the intersect function of R language to explore the differentially expressed chemokines in ovarian cancer,it was found that there were 27 differentially expressed chemokines,namely CCL2,CCL3,CCL4,CCL5,CCL7,CCL8,CCL11,CCL14,CCL20,CCL21,CCL25,CCL26,CCL27,CCL28,CXCL1,CXCL5,CXCL6,CXCL8,CXCL9,CXCL10,CXCL11,CXCL12,CXCL13,CXCL14,CXCL16,CXCL17,XCL2.4.Univariate Cox and Kaplan-Meier Plotter survival analysis of chemokines related to the prognosis of ovarian cancer found that a total of 6 chemokines related to the prognosis of ovarian cancer were CXCL13,CCL25,CXCL13,CCL8,CXCL9,CXCL11(P<0.05),the higher the expression level,the better the prognosis.5.Differentially expressed chemokines CCL25,CXCL13,CCL8,CXCL9,CXCL11,and CXCL10 related to the prognosis of ovarian cancer were included in the multivariate analysis to construct an ovarian cancer risk model.It was found that 4 chemokines including CXCL10,CXCL11,XCL13 and CCL25 finally participated in the model establishment.Using the model,patients with ovarian cancer were divided into high-risk and low-risk groups.The expression levels of 4 chemokines and the overall survival rate of patients in the high-risk group were significantly lower than those in the low-risk group(P<0.05).6.The receiver operating characteristic(ROC)curve was used to verify the accuracy of the risk model,and it was found that the area under the 1-year,3-year and 5-year ROC curve were 0.563,0.617,and 0.653,respectively,indicating that the model has a good predictive function.7.The CIBERSORT algorithm was used to calculate the proportion of 22 immune infiltrating cells in the high-risk and low-risk groups of ovarian cancer,and to analyze the correlation between CXCL10,CXCL11,CXCL13,CCL25 and various types of immune cells in ovarian cancer tissues.It was found that compared with ovarian cancer patients in the low-risk group,the number of monocytes,dendritic cells and neutrophils in the tissues of ovarian cancer patients in the high-risk group increased significantly.The four chemokines CXCL10,CXCL11,CXCL13,CCL25 and M1 type macrophages in the prognostic model of ovarian cancer are all positively correlated,the correlation coefficients are:0.558,0.581,0.469,0.539,and the P values are:2.04E-06,5.25E-05,4.47E-08,0.045,CCL25 is negatively correlated with M2 type macrophages,the correlation coefficient is-0.438,P=0.00045.CXCL13 is negatively correlated with mature dendritic cells,the correlation coefficient is-0.428,P=2.23E-05,CCL25,CXCL13 are negatively correlated with neutrophils,the correlation coefficients are:-0.358,-0.381,P=0.016,P=0.014.Conclusion:1.This study screened 6813 differentially expressed genes in ovarian cancer,and performed GO functional annotation and pathway analysis on differentially expressed genes,which can provide a theoretical basis for the research of ovarian cancer.2.Successfully excavated 27 differentially expressed chemokines in ovarian cancer,6 prognostic-related differentially expressed chemokines,and 4 chemokines involved in the construction of risk models were screened,and the 4 chemokines were used to construct a prognostic risk assessment for ovarian cancer Model,and at the same time evaluated the predictive value of the model,the model has a better predictive function.3.The number of mononuclear cells,dendritic cells and neutrophils in the tissues of the high-risk group of ovarian cancer patients increased significantly.CXCL10,CXCL11,CXCL13 and CCL25 are related to immune infiltrating cells in ovarian cancer tissues,and tend to Chemical factors may affect the occurrence and development of ovarian cancer through immune infiltrating cells,and affect the prognosis of patients.