基于生物信息学筛选早发性乳腺癌差异表达基因  

Screening for differentially expressed genes in early-onset breast cancer using bioinformatic analysis

作　　者：牟志公 谢明均 Mou Zhigong;Xie Mingjun(Department of Breast Surgery,Affiliated Hospital of Southuoest Medical Uniuersity,Luchou 64600,China)

机构地区：[1]西南医科大学附属医院乳腺外科,泸州646000

出　　处：《中华乳腺病杂志（电子版）》2022年第1期6-13,共8页Chinese Journal of Breast Disease(Electronic Edition)

基　　金：四川省科技厅应用基金重点研究项目(2017JY0029)。

摘　　要：目的:筛选并分析与早发性乳腺癌发生、发展相关的靶基因。方法:(1)在美国国立生物技术信息中心的公共基因芯片数据库(GEO)中检索早发性乳腺癌样本及非早发性乳腺癌样本相关基因芯片数据。对上述数据使用GEO2R、R4.1.2及Venn软件筛选出相关差异表达基因(DEGs),并运用在线分析工具(Web Gestalt)对DEGs,进行相关功能和信号通路富集分析。(2)同时,通过String在线数据库构建DEGs编码的蛋白质-蛋白质相互作用(PPI)网络,并利用Cytohubba插件对该网络中的基因进行评分,筛选出枢纽基因。将枢纽基因导入Kaplan-Meier生存分析工具(Kaplan-Meier Plotter),评估枢纽基因在早发性乳腺癌的预后价值。(3)将肿瘤基因组图谱(TCGA)数据库中的肿瘤组织以年龄为标准进行分组,分析枢纽基因在各年龄组中的表达,并与正常组织中的表达进行比较,对得到的枢纽基因进行验证。DEGs表达量的多组间比较使用Kruskal-Wallis H检验,使用Bonferroni法进行两两比较。结果:(1)筛选出编号为GSE89116、GSE109169、GSE36295的基因芯片数据集,共得到80个差异表达基因,其中上调差异表达基因17个,下调差异表达基因63个。富集分析显示:DEGs主要富集在脂质代谢和氧化还原过程以及PPAR信号通路、AMPK信号通路上。(2)在PPI中发现主要的关键基因为PPARG、ADIPOQ、LIPE、PCK1、PDK4、ACACB、PLIN1、CAV1、CD36、ANGPTL4。ACACB、ADIPOQ、CAV1、LIPE、PLIN1、PPARG基因的低表达与乳腺癌患者的不良OS相关(HR=0.69、0.84、0.76、0.88、0.78、0.82;95%CI:0.59~0.80、0.76~0.93、0.67~0.83、0.79~0.97、0.70~0.86、0.73~0.90;P均<0.050)。(3)ACACB、ADIPOQ、LIPE、PLIN1、CAV1及PPARG这6个与预后相关的基因在正常组织中的表达量均远高于各年龄组肿瘤组织中的表达量(χ^(2)=104.03、179.57、161.85、189.87、118.56、103.62,P均<0.001),早发性乳腺癌组(21~40岁)的LIPE、PLIN1表达量低于41~60岁、61~80岁年龄组,�Objective To search for and analyze the target genes associated with the ocurrence and development of early-onset breast cancer.Methods(1)The related gene microarray data on early-onset breast cancer and non-early-onset breast cancer were retrieved from the Gene Expression Omnibus(GEO)database of the U.S.National Center for Biotechnology Information(NCBI).The relevant difrentially expressed genes(DEGs)were screened for using GEO2R,R4.1.2 and Venn softwares,and the online analysis tool(Web Gestalt)was used to conduct the enrichment analysis of DEGs-related functions and signal pathways.(2)Meanwhile,a protein-protein interaction(PPI)network.was constructed from the String online database and Cytohubba plug-in was used to score the genes in the network.and screen out the key genes.Prognostic walue of key genes was assessed using the Kaplan-Meier online plotter.(3)Tumor tissues in the Cancer Genome Atlas(TCGA)database were grouped according to patient age,and the expression of keys genes in each age group was analyzed and compared with the expression in normal tssues to verify the obtained key genes.The multi-group companison of DEGs expression was performed using the Kruskal-Wallis H test,and Bonferroni method was used for pairwise comparison.I Results(1)Three gene microarray datasets(GSE89116,GSE109169 and GSE36295)were selected to obtain 80 DEGs,including 17 up-regulated genes and 63 down-regulated genes.Enrichment analysis showed that DEGs were mainly enriched in lipid metabolism,redox proces,PPAR signal pathway and AMPK signal pathway.(2)The main key genes found in PPI included PPARG,ADIPOQ,LIPE,PCK1,PDK4,ACACB,PLIN1,CAV1,CD36 and ANGPTL4.The low expressions of ACACB,ADIPOQ,CAV1,LIPE,PLIN1 and PPARG genes were associated with lower 0S in breast cancer patients(HR=0.69,0.84,0.76,0.88,0.78,0.82;95%CI:0.59-0.80,0.76-0.93,0.67-0.83,0.79-0.97,0.70-0.86,0.73-0.90;all P<0.050).(3)The expressions of six genes related to the prognosis(ACACB,ADIP0Q,LIPE,PLIN1,CAVI and PPARG)in normal tissues were sigificantly higher than

关 键 词：乳腺肿瘤 生物信息学 差异表达基因 

分 类 号：R737.9[医药卫生—肿瘤]