| 党永林,秦静,董瑞,等.基于 TCGA数据库构建胃癌免疫相关长链非编码 RNA预后模型[J].安徽医药,2026,30(6):1156-1161. |
| 基于 TCGA数据库构建胃癌免疫相关长链非编码 RNA预后模型 |
| Construction of an immune-related long non-coding RNA prognostic model for gastric cancer based on the TCGA database |
| |
| DOI：10.3969/j.issn.1009-6469.2026.06.019 |
| 中文关键词: 胃肿瘤 RNA，长链非编码 免疫相关基因 癌症基因组图谱数据库 风险模型 预后 |
| 英文关键词: Stomach neoplasms RNA, long noncoding Immune-related genes The Cancer Genome Atlas database Risk mod. el Prognosis |
| 基金项目:陕西省教育厅科学研究计划项目( 23JK0379)；西藏民族大学校内科研项目( 24MDY17) |
|
| 摘要点击次数: 50 |
| 全文下载次数: 14 |
| 中文摘要: |
| 目的运用生物信息学手段对癌症基因组图谱( TCGA)数据库中的胃癌转录组数据展开深入分析，以筛选和鉴定可能在胃癌发生发展中发挥重要作用的关键基因，并构建胃癌免疫相关长链非编码 RNA(lncRNA)预测模型。方法该研究起止时间为 2023年 12月至 2024年 8月。从 TCGA数据库下载胃癌组织基因数据及相应临床资料(数据下载截止日期为 2024年 1月 15日)对免疫相关 lncRNA进行单因素和多因素 Cox回归分析，构建胃癌相关免疫 lncRNA预后预测模型。根据中位风险评分值将病人，划分为高风险与低风险两组。通过 Kaplan-Meier生存曲线比较两组预后差异，依据曲线下面积( AUC)评估模型预测性能。从 GEO数据库中筛选数据集进行外部验证。通过 R软件绘制风险评分曲线图、生存分布散点图、免疫相关 lncRNA热图，采用 “scatterplot3d”包进行主成分分析，采用 “ggpubr”包绘制箱线图。通过 lncRNA模型结合胃癌病人临床病理特征，建预后列线图；分析胃癌免疫相关 lncRNA与临床病理特征之间的关联；利用基因集富集分析( GSEA)软件包对免疫基因集进构行富集分析。结果通过多因素 Cox回归分析，以最小化赤池信息量准则( AIC)值为目标，通过双向逐步回归迭代筛选，最终得到 6个胃癌病人预后相关免疫 lncRNA(AL391244.1、AL513123.1、AC110995.1、REPIN1-AS1、SERPINB9P1、AC245041.2)，其中 REPIN1-AS1［HR=0.77，95%CI:(0.60，0.99)P=0.042］、 AC245041.2［HR=1.18，95%CI:(1.03，1.35)P=0.017］与胃癌免疫预后相关，基于上述6个lncRNA构建胃癌免疫相关l，ncRNA预后风险预测模型。外部验证提示模型具有稳，定预测效能；经多变量 Cox回归系数计算每例病人风险评分，根据评分的中位风险值，将病人分为低、高风险两组。生存曲线表明，低风险组病人生存率显著高于高风险组( P<0.05)。预后风险模型的 AUC为 0.686，显示其具备良好效能。结合胃癌病人的临床病理特征与 ln. cRNA模型，多因素 Cox回归分析结果表明，年龄和 lncRNA模型为胃癌预后的独立危险因素( P<0.05)。通过 R软件 “ggpubr”包绘制箱线图，结果提示基因 SERPINB9P1、AC245041.2与临床分期相关( P<0.05)AC110995.1、SERPINB9P1与 T分期相关( P<0.05)，AL513123.1与 N分期相关( P<0.05)，REPIN1-AS1与 M分期相关( P<0.05)免疫相关 lncRNA预后风险模型为胃。结论，癌病人预后的独立危险因素，可以较好地预测胃癌病人的预后情况，指导个体化治疗，可作为胃癌独立的预后生物标志物。 |
| 英文摘要: |
| Objective This study aimed to employ bioinformatics approaches to conduct an in-depth analysis of gastric cancer (GC)transcriptomic data from The Cancer Genome Atlas (TCGA) database, with the goal of identifying key genes potentially critical in thepathogenesis and progression of GC. Additionally, we sought to construct a prognostic prediction model for immune-related long non-coding RNAs (lncRNAs) in GC.Methods The study was conducted from December 2023 to August 2024. Gene expression data andcorresponding clinical information of gastric cancer tissues were downloaded from the TCGA database (data were downloaded as of Jan.uary 15, 2024). Univariate and multivariate Cox regression analyses were performed on immune.related lncRNAs, and a prognostic pre.diction model for gastric cancer.associated immune lncRNAs was constructed.Patients were stratified into high-and low-risk groups based on the median risk scores. Kaplan-Meier survival curves were generated to compare prognosis between groups, and the model'spredictive performance was evaluated using the area under the curve (AUC). External validation was conducted using datasets from theGene Expression Omnibus (GEO) database. R software was utilized to generate risk score curves, survival distribution scatter plots,heatmaps of immune-related lncRNAs, principal component analysis (PCA) plots (via the "scatterplot3d" package), and boxplots (viathe "ggpubr" package). A prognostic nomogram integrating the lncRNA model with clinicopathological features of GC patients was con.structed. Associations between immune-related lncRNAs and clinicopathological characteristics were analyzed, and gene set enrich.ment analysis (GSEA) was performed on immune-related gene sets.Results Through multivariate Cox regression analysis with the ob. jective of minimizing the Akaike information criterion (AIC), six immune-related lncRNAs associated with the prognosis of gastric can.cer patients were selected by bidirectional stepwise regression iteration: AL391244.1, AL513123.1, AC110995.1, REPIN1-AS1, SER. PINB9P1, and AC245041.2. Among these, REPIN1-AS1 [HR=0.77, 95% CI: (0.60, 0.99), P=0.042] and AC245041.2 [HR=1.18, 95% |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
