基于生物信息学分析筛选肺腺癌免疫预后相关lncRNA

Screening of lncRNA Related to Immune Prognosis of Lung Adenocarcinoma Based on Bioinformatics Analysis

目的基于TCGA和Molecular Signatures Database数据,利用生物信息学方法鉴定潜在的免疫预后相关的lncRNA(IRL)生物标志物,用于肺腺癌患者的预后预测。方法从TCGA获取肺腺癌IRL数据集和相应的临床参数;从Molecular Signatures Database v7.0中提取免疫基因。通过构建免疫基因共表达网络鉴定免疫相关的IRL。随后对IRL进行Cox回归分析、Lasso回归分析筛选肺腺癌预后相关的IRL生物标志。通过生存分析、ROC曲线分析评估IRL预测患者生存预后的价值。结果共鉴定了1124个IRL,通过单因素Cox回归分析,初步筛选出18个和肺腺癌预后相关的IRL,且通过Lasso回归分析对预后相关的IRL进行二次选择后得到14个IRL与肺腺癌的生存预后相关。Cox分析筛选出其中8个IRL作为预后的独立危险因子用于构建风险评分模型。K-M生存分析显示,低风险组生存时间长于高风险组(P<0.001)。ROC曲线分析显示,风险评分、年龄、性别、分期、T、M、N对肺腺癌预后的曲线下的面积分别为0.827、0.498、0.579、0.733、0.673、0.508、0.685。对风险模型绘制风险曲线,结果显示随着风险值的升高,患者生存时间逐渐下降,死亡人数逐渐增多。结论IRL构建的8个风险模型可能是肺腺癌患者潜在的预测预后生物标志物。

Objective To identify potential immunoprognostic lncRNA(IRL)biomarkers by bioinformatics methods based on TCGA and Molecular Signatures Database data for prognosis prediction of lung adenocarcinoma patients.Methods The IRL dataset of lung adenocarcinoma and corresponding clinical parameters were obtained from TCGA.Immune genes were extracted from Molecular Signatures Database v7.0.Immune-related IRL was identified by constructing an immune gene co-expression network.Cox regression analysis and Lasso regression analysis were used to screen IRL biomarkers related to the prognosis of lung adenocarcinoma.The value of IRL in predicting the survival prognosis of patients was evaluated by survival analysis and ROC curve analysis.Results A total of 1124 IRLs were identified.Through univariate Cox regression analysis,18 IRLs related to the prognosis of lung adenocarcinoma were initially screened,and 14 IRLs related to the survival and prognosis of lung adenocarcinoma were obtained after secondary selection of prognosis-related IRLs by Lasso regression analysis.Cox analysis screened 8 IRLs as independent risk factors for prognosis to construct a risk scoring model.K-M survival analysis showed that the survival time of the low-risk group was longer than that of the high-risk group(P<0.001).ROC curve analysis showed that the area under the curve of risk score,age,gender,stage,T,M and N for the prognosis of lung adenocarcinoma was 0.827,0.498,0.579,0.733,0.673,0.508 and 0.685,respectively.The risk curve was drawn for the risk model.The results showed that with the increase of the risk value,the survival time of the patients gradually decreased,and the number of deaths gradually increased.Conclusion Eight risk models constructed by IRL may be potential prognostic biomarkers for patients with lung adenocarcinoma.