标题：基于TCGA数据库分析乳腺癌中TERT相关分子网络机制

      作者：陈运景 1，何贵省 2*，苏亚静 1，吴灿章 1，敬波 1，吴煌福 2    1.海南医学院研究生院，海南 海口 570100；2.海南医学院第二附属医院乳甲外科，海南 海口 570100

      卷次： 2021年32卷19期

      【摘要】 目的 通过TCGA数据库中的数据分析TERT基因在乳腺恶性肿瘤中的表达及生物学作用。方法 从癌症基因组图谱(TCGA)数据库下载乳腺癌的基因表达数据与临床资料。利用R软件包进行TERT表达差异分析及其与患者预后相关性；通过共表达网络及LinkedOmics在线数据库分析共表达基因及潜在调控mRNA分子。使用Spearman和CIBERSORT算法，分析TERT与肿瘤微环境的相关性。使用GDSC在线数据库分析TERT基因与常见化疗药物的敏感性。通过GSEA富集分析TERT可能参与的信号通路。结果 TERT在乳腺癌样本中的表达水平显著上调，差异有统计学意义(P<0.05)，其表达量与患者的 stage分期呈显著正相关(P<0.05)。通过单、多因素Cox分析表明TERT基因可作为乳腺癌的独立预后因子[P<0.05，HR=1.474 (1.138~1.909)]。通过共表达及LinkedOmics数据库分析结果表明 TERT表达与 ALKAL1、CNPY1、FSD1、GF1B、HMGB1P1等 5个基因正相关及与 DERA、MCM5、AC004858.1、SLCTA14、CACNA1G-AS1等 5个基因呈负相关，与mir-1245、mir-337、mir-10b、mir-199a-1、mir-377等5个miRNA呈正相关。CIBERSORT算法和Spearman相关系数分析表明TERT与T cells CD4 memory activated、Macrophages M0、Macrophages M1显著正相关，与Mast cells resting、T cells CD4 memory resting显著负相关。通过GDSC数据库分析表明TERT的表达与达沙替尼、吉西他滨、埃罗替尼以及伊马替尼、吉非替尼、顺铂等药物的敏感性相关(P<0.05)。TERT可富集在核糖核酸聚合酶、错配修复、原发性免疫缺陷信号通路。结论 TERT基因在乳腺癌样本中高表达及其与患者预后较差有关，表明TERT基因是乳腺癌的发病机制、诊断及治疗的靶点。
      【关键词】 TERT；乳腺恶性肿瘤；相关分子网络；生物信息学
      【中图分类号】 R737.9 【文献标识码】 A 【文章编号】 1003—6350（2021）19—2454—08

Analysis of the network mechanism of TERT related molecular in breast cancer based on TCGA database.
CHENYun-jing 1, HE Gui-xing 2*, SU Ya-jing 1, WU Can-zhang 1, JING Bo 1, WU Huang-fu 2. 1. Graduate School, HainanMedical University, Haikou 570100, Hainan, CHINA; 2. Breast Armor Surgical Outpatient, the Second Affiliated Hospital ofHainan Medical University, Haikou 570100, Hainan, CHINA
【Abstract】 Objective To explore the expression and biological effects of TERT gene in malignant breast tu-mor through the TCGA database. Methods The gene expression data and clinical data of breast cancer were download-ed from the Cancer Genome Atlas (TCGA) database. R software package was used to analyze the difference of TERT ex-pression and its correlation with patient's prognosis. Co-expression genes and potentially regulated mRNA moleculeswere analyzed through the co-expression network and LinkedOmics online database. Spearman and CIBERSORT algo-rithm were used to analyze the correlation between TERT and tumor microenvironment. GDSC online database was usedto analyze the sensitivity of TERT gene and common chemotherapy drugs. The possible signaling pathways of TERTwere analyzed through GSEA enrichment. Results The expression level of TERT in breast cancer samples was signifi-cantly increased with statistically significant difference (P<0.05); its expression was significantly positively correlatedwith the stage staging of patients (P<0.05). Single and multivariate Cox analysis showed that TERT gene could be usedas an independent prognostic factor for breast cancer (P<0.05, HR=1.474 [1.138-1.909]). Co-expression and LinkedO-mics database analysis showed that TERT expression was positively correlated with 5 genes, including ALKAL1,CNPY1, FSD1, GF1B, and HMGB1P1; negatively correlated with 5 genes, including DERA, MCM5, AC004858.1,SLCTA14, and CACNA1G-AS1; and positively correlated with 5 miRNAs, including mir-1245, mir-337, miR-10b,mir-199a-1, and mir-377. CIBERSORT algorithm and Spearman correlation coefficient analysis showed that TERT wassignificantly positively correlated with T cells CD4 memory activated, Macrophases M0, Macrophases M1, and signifi-cantly negatively correlated with Mast cells resting and T cells CD4 memory resting. GDSC database analysis showed thatTERT expression was associated with the sensitivity of dasatinib, gemitabine, erotinib, imatinib, giefetinib, and cisplatindoi:10.3969/j.issn.1003-6350.2021.19.002基金项目：海南省重点研发计划项目(编号：ZDYF2017087)