目的 通过生物信息学方法研究与铁死亡相关的治疗胆管癌潜在作用机制，高通量筛选通过铁死亡途径治疗胆管癌疾病的潜在小分子药物。方法 在GEO（Gene Expression Omnibus）在线数据库中检索与胆管癌相关的数据集，使用TCGA（The Cancer Genome Atlas）数据库获得胆管癌相关样本及正常样本的数据，将上述筛选得到的数据集矩阵进行规范化数据整理，通过Sanger平台对数据集中的靶点基因进行limma差异分析，通过FerrDb铁死亡相关分析数据库收集与铁死亡相关的靶点，将收集得到的铁死亡相关靶点与差异分析得到的胆管癌相关疾病靶点取交集后，得到共同靶点，将上述共同靶点导入STRING数据库，得到与铁死亡相关的胆管癌疾病靶点间的蛋白质相互作用（PPI）网络分析，筛选核心靶点，分别使用DAVID v6.8与Funrich富集分析软件对上述潜在作用靶点进行生物学功能与信号通路的富集分析，并利用cMAP（Connectivity Map）数据库高通量筛选通过铁死亡途径治疗胆管癌疾病的潜在小分子药物，将小分子化合物与靶蛋白进行分子对接分析，通过TIMER（Tumor Immune Estimation Resource）数据库分析铁死亡与胆管癌疾病共同靶点各自间表达水平与免疫细胞浸润水平或免疫检查点表达水平的相关性。结果 筛选得到3 133个胆管癌相关靶点，铁死亡相关靶点487个，取交集处理共得到93个铁死亡与胆管癌疾病相关的共同靶点，共同靶点的基因本体（GO）功能富集分析结果提示铁死亡途径对胆管癌治疗方面可能与转录调控相关过程、凋亡与细胞增殖等密切相关，京都基因与基因组百科全书（KEGG）信号通路富集分析结果表明铁死亡途径对胆管癌治疗方面可能与TRAJL分子信号通路、表皮生长因子受体（EGFR/ErbB）相关信号通路、鞘氨醇-1-磷酸受体-1（S1P1）分子信号通路、胰岛素相关信号通路等过程密切相关，通过筛选得到3个小分子化合物（海恩酮、氯可托龙戊酸酯、利扎曲普坦），分子对接结果表明氯可托龙戊酸酯与其他2种成分相比，对上述3个潜在核心共同作用靶点（TP53、PTEN、SRC）的结合普遍较其余小分子成分强，从而提示其有望通过铁死亡途径治疗胆管癌相关疾病。结论 以铁死亡途径作为切入点，通过生物信息学方法探寻治疗胆管癌的分子作用机制及蛋白靶点，从而筛选得到潜在的治疗药物，具有重要意义。
Objective To screen the potential therapeutic targets of cholangiocarcinoma associated with ferroptosis using bioinformatics methods, and to high-throughput screen the potential small molecule drugs for the treatment of cholangiocarcinoma diseases through ferroptosis mechanisms. Methods Data sets related to cholangiocarcinoma were retrieved from GEO (Gene Expression Omnibus) online database, and the data of cholangiocarcinoma related samples and normal samples were obtained from TCGA (The Cancer Genome Atlas) database. The data set matrix obtained from the above screening was standardized, and limma difference analysis was performed on target genes in the data set through Sanger platform, and targets related to ferroptosis were collected through FerrDb ferroptosis related online database, and after intersecting the collected ferroptosis related targets with cholangiocarcinoma targets obtained through differential analysis methods, a common target is obtained. The common targets are imported into the STRING database to obtain protein-protein interaction (PPI) network analysis between ferroptosis related cholangiocarcinoma targets. Core targets are screened, and biological functional and signaling pathway enrichment analysis are performed on the above potential targets using DAVID v6.8 and Funrich enrichment analysis software. Using the cMAP (Connectivity MAP) database to high-throughput screen potential small molecule drugs for the treatment of cholangiocarcinoma through the ferroptosis pathway, and molecular docking analysis was performed on small molecule compounds and target proteins. The TIMER (Tumor Immune Estimation Resource) database was used to analyze the correlation between the expression levels of common targets of ferroptosis and cholangiocarcinoma, immune cell infiltration levels, or immune checkpoint expression levels. Results 3 133 targets related to cholangiocarcinoma and 487 targets related to ferroptosis were identified, and a total of 93 common targets related to ferroptosis and cholangiocarcinoma were obtained by intersection. The GO functional enrichment analysis of the common targets suggested that the ferroptosis pathway may be closely related to the treatment of cholangiocarcinoma. The KEGG signaling pathway enrichment analysis results showed that the ferroptosis pathway may be related to the TRAJL molecular signaling pathway in the treatment of cholangiocarcinoma. The EGF(ErbB) receptor related signaling pathway, S1P1 molecular signaling pathway, insulin related signaling pathway, and other processes are also closely related it. Meanwhile, some small molecule compounds were identified through screening, and a total of three components were identified. Molecular docking results showed that compared with the other two components, the binding of chlorotolone valerate to the three potential core targets mentioned above is generally stronger than the other small molecule components, this suggests that it has the potential to treat cholangiocarcinoma related diseases through the ferroptosis pathway. Conclusion Using the ferroptosis pathway as a starting point, potential targets for cholangiocarcinoma can be identified through bioinformatics methods, and potential therapeutic drugs can be screened.