基于生物信息学发掘与鉴定食管癌潜在生物标志物
Identification of Potential Biomarkers for Esophageal Cancer Based on Bioinformatics

目的：基于生物信息学，对食管癌(Esophageal cancer, ESCA)潜在的生物标志物在食管癌发生发展中的作用机制进行探究。方法：通过GEO数据库(https://www.ncbi.nlm.nih.gov/geo/)和TCGA数据库 (https://portal.gdc.cancer.gov)获取食管癌数据集后，按照p < 0.05的标准进行差异基因的筛选。将初步筛选后的基因通过四种机器学习方法(Boruta、Lasso、XGBoost、随机森林和SVM-RFE算法)对筛选出的关键基因在食管癌中的重要性进行评估后，对其进行通路分析用以揭示其作用机制。进行受试者工作特征曲线(receiver operating characteristic curves analysis, ROC)分析，评估基因对于食管癌的诊断效能。最后对基因进行生存分析与单细胞分析，分别评估基因对ESCA预后的影响与基因在肿瘤组织中单个细胞水平的表达情况。绘制了ADAM12、CTHRC1、IBSP和OLR1基因在食管癌中的多组学图谱，揭示了其对于食管癌的发生、发展与预后的高度相关性。结果：ADAM12、CTHR1、IBSP、OLR1为食管癌的危险因素，其表达的高低对于预测食管癌的发生和预后具有很高的准确性，并且低表达组相比于高表达组预后更加良好。结论：得到了与食管癌的发生和预后相关的四个基因：ADAM12、CTHR1、IBSP、OLR1，可能成为食管癌早期诊断的生物标志物。
Objective: To investigate the mechanism of potential biomarkers in the occurrence and development of esophageal cancer (ESCA) based on bioinformatics methods. Methods: Esophageal cancer-related datasets were retrieved from the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) database and The Cancer Genome Atlas (TCGA, https://portal.gdc.cancer.gov) database. Differentially expressed genes were screened with the criterion of p < 0.05. The importance of the initially screened genes in esophageal cancer was evaluated by four machine learning methods, namely Boruta algorithm, Lasso algorithm, extreme gradient boosting (XGBoost) algorithm, random forest algorithm, and support vector machine-recursive feature elimination (SVM-RFE) algorithm. Pathway analysis was subsequently performed to elucidate their underlying mechanisms of action. Receiver operating characteristic (ROC) curve analysis was conducted to assess the diagnostic efficacy of these genes for esophageal cancer. Finally, survival analysis and single-cell analysis were carried out to determine the impact of the target genes on the prognosis of ESCA patients and their expression characteristics at the single-cell level in tumor tissues, respectively. Multi-omics profiles of ADAM12, CTHRC1, IBSP and OLR1 genes in esophageal cancer were plotted, and their high correlation with the occurrence, development and prognosis of esophageal cancer was analyzed. Results: ADAM12, CTHRC1, IBSP and OLR1 were identified as risk factors for esophageal cancer. The expression levels of these genes showed high accuracy in predicting the occurrence and prognosis of esophageal cancer, and patients in the low-expression group had significantly better prognosis than those in the high-expression group. Conclusion: Four genes (ADAM12, CTHRC1, IBSP, OLR1) closely associated