结肠腺癌能量代谢相关mRNA-LncRNA预后预测模型的构建与验证

Construction and validation of prognostic prediction model of energy metabolism related mRNA-LncRNA in colon adenocarcinoma

目的构建和验证能量代谢相关信使RNA(mRNA)-长链非编码RNA(LncRNA)在结肠腺癌(COAD)患者中的预后预测模型。方法从癌症基因组图谱(TCGA)中下载TCGA-COAD项目中COAD转录本数据与临床病理信息,转录本数据用于筛选差异表达的mRNA(DEGs)与差异表达的LncRNA(DELs)。从MSigDB数据库中收集能量代谢相关基因,与DEGs取交集得到能量代谢相关DEGs,再将其与DELs进行共表达分析得到能量代谢相关DEGs、DELs。将得到的能量代谢相关DEGs与DELs纳入Cox回归模型中,筛选并建立COAD患者的预后预测模型。根据预测模型结果建立风险得分,将其分为高风险组与低风险组,然后利用Kaplan-Meier曲线、受试者工作特征曲线(ROC)、Cox回归分析对风险得分进行评估。结果共表达分析得到125个能量代谢相关DEGs和451个能量代谢相关DELs。Cox回归结果显示1个mRNA(LRP2)与5个LncRNA(分别为CTC-428G20.6、LINC02257、PRR7-AS1、RP11-29G8.3、RP11-677M14.3)与COAD患者预后独立相关。Kaplan-Meier分析显示高风险组与低风险组的中位数生存期分别是5.066年和8.334年(P<0.05);1、3、5年生存的患者风险得分的ROC曲线下面积分别是0.767、0.694和0.691;多因素Cox回归显示高风险组生存时间明显短于低风险组(HR=3.78,95%CI:2.29~6.22,P<0.05)。结论研究成功构建COAD能量代谢相关mRNA-LncRNA预后预测模型,该模型具有较好的预测性能。

Objective To construct and verify the prognostic prediction model of energy metabolism related mRNA-LncRNA in the patients with colon adenocarcinoma(COAD).Methods The transcripts and clinicopathological information of TCGA-COAD project were downloaded from TCGA database.The transcripts were used to screen differentially expressed mRNA(DEGs)and differentially expressed LncRNA(DELs).The energy metabolism related genes were collected from MSigDB database,and the energy metabolism-related DEGs were obtained by intersection with DEGs,and then the energy metabolism-related DEGs-DELs were obtained by co expression analysis with metabolism-related DEGs and DELs.The obtained energy metabolism related DEGs and DELs were included in the Cox regression model to screen and establish the prognosisprediction model of the patients with COAD.According to the results of the prediction model,the risk score was established and divided into the high-risk group and low-risk group.Then the Kaplan Meier curve,receiver operating characteristic(ROC)curve and Cox regression analysis were used to evaluate the risk score.Results The co-expression analysis obtained 125 energy metabolism-related DEGs and 451 energy metabolism-related DELs.The Cox regression results showed that one mRNA:LRP2 and five LncRNAs(CTC-428G20,LINC02257,PRR7-AS1,RP11-29G8.3,RP11-677M14.3)were independently correlated with the prognosis of COAD patients.The Kaplan Meier analysis showed that the median survival time of the high and low risk groups were 5.066 years and 8.334 years respectively(P<0.05);the area under the ROC curve of risk score for 1-,3-,and 5-year survival was 0.767,0.694,and 0.691,respectively;the multivariate Cox regression showed that the survival time of the high-risk group was significantly lower than that of the low-risk group(HR=3.78,95%CI:2.29-6.22,P<0.05).Conclusion This study successfully constructs the prognosis prediction model of energy metabolism related mRNA-LncRNA in COAD,which has good predictiv performance.