【摘要】 目的:通过生物信息学分析构建肺腺癌蛋白质预后模型。方法:从癌症蛋白质组图谱和肿瘤基因组图谱数据库分别获取肺腺癌的蛋白质数据和相应的临床数据，通过单因素Cox回归分析和逐步回归分析筛选与肺腺癌预后相关的蛋白质，建立预后风险模型；使用多因素Cox回归对其进行预后风险评分分析，计算曲线下面积(AUC)评价模型的稳健性和准确性。结果:筛选出3个与生存显著相关的蛋白质用于预后模型的构建；预后模型风险评分与预后显著相关(P<0.001),可作为评估患者预后的独立风险因子；预后模型AUC=0.710,说明模型具有稳定的特异性和灵敏度。结论:该预后模型能准确预测肺腺癌患者的总体生存率，有助于临床早期识别预后不良的肺腺癌患者并对其进行早期干预治疗，对提高肺腺癌的生存率具有重要意义。此外，筛选出3个促进肺腺癌进展的风险蛋白有望成为肺腺癌治疗的新靶点。更多还原

【Abstract】 Objective:To establish the protein prognostic model for lung adenocarcinoma based on bioinformatics. Methods:The protein data and related clinical data of lung adenocarcinoma were obtained from The Cancer Proteome Atlas(TCPA) and The Cancer Genome Atlas(TCGA) database respectively, and the proteins associated with the prognosis of lung adenocarcinoma were screened using univariate Cox regression analysis and stepwise regression to establish the prognostic risk model. The model was analyzed by prognostic risk score using multivariate Cox regression, and the area under the curve(AUC) was calculated to evaluate the robustness and accuracy of the model. Results:Three proteins significantly correlated with survival were screened for the construction of the prognostic model. The risk score of the prognostic model was significantly correlated with the prognosis(P<0.001), which could be used as an independent risk factor to evaluate the prognosis of patients. The AUC of the prognostic model was 0.710, indicating that the model had stable specificity and sensitivity. Conclusion:The prognosis model could accurately predict the overall survival rate of patients with lung adenocarcinoma, which is helpful for early identification of patients with poor prognosis of lung adenocarcinoma and early intervention treatment, and is of great significance to improve the survival rate of patients with lung adenocarcinoma. In addition, three risk proteins screened in this study that promote the progression of lung adenocarcinoma are expected to become new targets for the treatment of lung adenocarcinoma.更多还原