betway必威登陆网址 (betway.com )学报››2023,Vol. 44››Issue (11): 819-825.DOI:10.3969/j.issn.2097-0005.2023.11.004

• 基础研究 •上一篇

基于生物信息学预测影响卵巢癌患者总体生存率的铁死亡相关基因

陈治宇1(), 徐晗2, 于德新1()

  1. 1.山东大学齐鲁医院放射科,山东 济南 250012
    2.济南市第三人民医院影像中心,山东 济南 250132
  • 收稿日期:2023-09-04出版日期:2023-11-25发布日期:2024-01-22
  • 通讯作者:于德新
  • 作者简介:陈治宇,学士,初级技师,研究方向:医学影像技术应用及数据处理,E-mail:Chenzy581X@163.com

Prediction of ferroptosis-related genes affecting overall survival in ovarian cancer patients based on bioinformatics

Zhiyu CHEN1(), Han XU2, Dexin YU1()

  1. 1.Department of Radiology,Qilu Hospital of Shandong University,Jinan 250012,China
    2.Department of Medical Imaging,Jinan Third People's Hospital,Jinan 250132,China
  • Received:2023-09-04Online:2023-11-25Published:2024-01-22
  • Contact:Dexin YU

摘要:

目的通过生物信息学分析筛选影响卵巢癌(ovarian cancer, OC)患者总体生存率的铁死亡相关基因(ferroptosis-related genes, FRGs)。方法通过癌症基因组图谱公共数据库(the cancer genome altas, TCGA) 和基因组织表达数据库(genotype-tissue expression, GTEx)下载OC患者的mRNA表达谱和相应的临床数据,使用铁死亡数据库(ferroptosis database, FerrDb)获得铁死亡相关基因,使用R软件确定具有差异表达和预后价值的铁死亡相关基因,并在此基础上使用Lasso回归分析构建风险模型。通过受试者工作特征曲线(receiver operator characteristic curve, ROC)评估模型对预后的准确性,使用单因素和多因素分析筛选独立预后因素。应用功能富集和单样本基因富集分析(single-sample gene set enrichment analysis, ssGSEA)探索潜在机制。结果最终由18个基因构建了预后模型,并将OC患者分为高风险和低风险组。与低风险组相比,高风险组OC患者的总生存期(overall survival, OS)显著。时间依赖ROC曲线显示第1,2,3年的曲线下面积(area under curve, AUC)分别为0.664,0.702和0.675,说明风险模型较为可靠。风险评分被证明是OS的独立预测因子(HR= 4.092, 95%CI:2.862 ~ 5.852,P< 0.001)。功能分析显示,差异可能与受体和配体的相互作用、细胞因子和细胞因子受体的相互作用有关。抗原呈递过程包括活化的树突状细胞(activate dendritic cells, aDCs)、浆细胞样树突状细胞(plasmacytoid dendritic cells, pDCs)、抗原呈递细胞(antigen presenting cells, APC)共抑制、主要组织相容性复合体(major histocompatibility complex, MHC) I类评分和人白细胞抗原(human leukocyte antigen, HLA),差异均具有统计学意义(调整后P< 0.05)。结论FRGs影响OC患者的总体生存率,为预测OC的预后及探索新的治疗方式提供了新的方向,但FRGs与肿瘤免疫之间的潜在机制仍值得进一步研究。

关键词:卵巢癌,铁死亡,生物信息学,总生存期,免疫状态

Abstract:

ObjectiveTo screen ferroptosis-related genes (FRGs) that affect the overall survival rate of patients with ovarian cancer by bioinformatics analysis.MethodsThe mRNA expression profiles and corresponding clinical data of ovarian cancer (OC) patients were downloaded from TCGA and GTEx public databases, ferroptosis-related genes were obtained using FerrDb database, FRGs with differential expression and prognostic value were identified by R software, and on this basis Lasso regression analysis was used to construct risk models. The accuracy of the model for prognosis was assessed by receiver operator characteristic curve (ROC), and independent prognostic factors were screened using univariate and multivariate analysis. Functional enrichment and single-sample gene set enrichment analysis (ssGSEA) were applied to explore potential mechanisms.ResultsA prognostic model was finally constructed from 18 genes, and OC patients were divided into high-risk and low-risk groups. Compared with the low-risk group, the overall survival (OS) of OC patients in the high-risk group was significantly worse. The time-dependent ROC curve showed that the area under the curve (AUC) of the first, second, and third years were 0.664, 0.702 and 0.675, respectively, indicating that our risk model was more reliable. Risk score was shown to be an independent predictor of OS (HR= 4.092, 95%CI:2.862 ~ 5.852,P< 0.001). Functional analysis showed that the differences might be related to the interaction of receptors and ligands, and the interaction of cytokines and cytokine receptors. Antigen presentation process including aDCs, pDCs, APC co-suppression, MHC class I score and HLA were all significantly different (adjustedP< 0.05).ConclusionFRGs affect the overall survival rate of ovarian cancer patients, which provides a new direction for predicting the prognosis of OC and exploring new treatment methods, but the underlying mechanism between FRGs and tumor immunity needs further study.

Key words:ovarian cancer,ferroptosis,bioinformatics,overall survival,immune status