Objective To construct a prognostic model for patients with endometrial cancer(EC)by bioin-formatics.Methods To download EC-related prognostic data from the TCGA database,the data downloaded from the TCGA were divided into Train group and Test group.A prognostic model was established using data from the Train group,and whether the prognostic model had a good predictive ability was verified in the Test group.Results 229 genes highly related to survival were screened by single-factor Cox regression,and 16 genes with specific significance were screened for establishing the risk model.Using random forest survival(ran-domsurvivalforestsalgorithm)algorithm 9 secondary screening of the genes associated with survival to risk prog-nostic model,MGAT4A,SRD5A1,MBOAT2,EPHA10,L1CAM,CHODL,AC010729.1,RP11-66D17,RP5-1158E12.3 are all considered to be risk effect factors of HR＞1(P＜0.05).The calculation formula of risk score is as follows:risk score=∑iwixi+1.802659.These samples were evaluated for high and low risk by risk score evaluation formula.K-M survival curves were drawn in the Train and Test groups.The results show that the overall survival rate of the high-risk group was lower than that of the low-risk group(P＜0.05).Conclusion(1)Patients with endometrial cancer were divided into high and low risk groups by the score calcu-lated by the risk model,and the difference in survival status between the two groups was statistically significant(P＜0.001).(2)This prognostic risk model can effectively predict the prognosis of EC patients and provide a new basis for the clinical diagnosis and treatment of EC.
the cancer genome atlasendometrial carcinomaprognostic modelbioinformatics
维普
万方数据
