摘要
目的基于TCGA数据库筛选影响前列腺癌(PCa)风险水平的关键基因,并建立PCa患者生存风险预测模型。方法从TCGA数据库下载PCa患者基因表达数据及相关临床数据,通过前期研究初步筛选基因,并将患者分为高、低风险两类;对基因进行差异表达分析和GO和KEGG通路富集分析,筛选相关基因和信号通路;对差异表达基因进行蛋白互作网络分析,标记出关键基因;将关键基因的表达数据与PCa患者生存时间纳入Cox回归分析,建立生存风险预测模型。结果前期研究得到620个基因,高风险患者234例,低风险患者285例;差异表达分析获得30个基因,主要分子功能(MF)为:受体结合和生长因子活动,生物学过程(BP)主要为细胞-细胞信号传导、细胞增殖的积极调节、血管生成的调节和细胞表面受体信号通路,细胞组分(CC)主要定位于细胞外区域,而KEGG信号通路为细胞因子-细胞因子受体相互作用;蛋白互作分析中共7个基因有相互作用,Cytoscape筛选出5个关键基因:PHYHIPL、CNTFR、GFRA1、EDN3和PROK1。结论通过本研究识别的影响PCa预后的关键基因,发现潜在的PCa风险靶点,可能为PCa的治疗和预后提供帮助。
Objective To screen the key genes affecting prostate cancer(PCa)risk level based on the TCGA database and establish a survival risk prediction model for PCa patients.Methods Download PCa patient gene expression data and related clinical data from the TCGA database,preliminary screening of genes through early research,and classify patients into high and low risk categories;perform differential expression analysis of genes and enrichment analysis of GO and KEGG pathways to screen relevant gene and signal pathway;Perform protein interaction network analysis on differentially expressed genes to mark key genes;incorporate expression data of key genes and PCa patient survival time into Cox regression analysis to establish a survival risk prediction model.Results 620 genes were obtained in previous studies,234 patients were high-risk patients,285 patients were low-risk patients;30 genes were obtained by differential expression analysis.The main molecular functions(MF)were:receptor binding and growth factor activity,and the main biological process(BP)For cell-cell signaling,positive regulation of cell proliferation,regulation of angiogenesis,and cell surface receptor signaling pathways,the cell component(CC)is mainly located in the extracellular region,while the KEGG signaling pathway is a cytokine-cytokine receptor interactions:A total of 7 genes interacted in the protein interaction analysis.Cytoscape screened out 5 key genes:PHYHIPL,CNTFR,GFRA1,EDN3,and PROK1.Conclusion The key genes affecting the prognosis of PCa identified through this study,and the discovery of potential PCa risk targets may help the treatment and prognosis of PCa.
作者
李熹阳
谷明宇
华琳
LI Xi-yang;GU Ming-yu;HUA Lin(School of Public Health,School of Bioengineering,Beijing 100069,China;School of Public Health,Capital Medical University,Beijing 100069,China)
出处
《医学信息》
2020年第2期80-85,共6页
Journal of Medical Information
关键词
前列腺癌
基因差异表达分析
生物信息学
富集分析
Prostate cancer
Differential gene expression analysis
Bioinformatics
Enrichment analysis
