基于生物信息学方法筛选结直肠癌转移的相关基因

Identification of related genes associated with colorectal cancer metastasis based on bioinformatics analysis

摘要:

目的： 应用生物信息学对基因芯片数据进行分析，探讨结直肠癌的转移机制，寻找潜在的转移及预后标记物。方法： 从GEO数据库选取GSE41568、GSE68468进行分析，使用R语言limma包筛选结直肠原发癌与转移瘤之间的差异基因，获得2个数据集的共同差异基因；运用clusterProfiler包进行功能富集分析并进行可视化；通过STRING数据库、Cytoscape软件进行蛋白质互作网络分析及关键模块的筛选；使用TCGA数据库的结直肠癌数据对差异基因进行表达分析和生存分析。结果： 共筛选出108个共同差异基因；通过基因富集分析发现，差异基因主要富集在补体及凝血级联等通路及生物学过程中；通过蛋白质互作网络分析发现3个关键模块；基于TCGA数据对共同差异基因分析发现，CLCA1、COLEC11、FCGBP、PDZD2、SERPINA1、SPINK4共6个基因在Ⅰ-Ⅱ和Ⅲ-Ⅳ期结直肠癌的表达有显著差异（P<0.05），并且与结直肠癌的预后显著相关（P<0.05）。结论： 通过生物信息学筛选出108个共同差异基因及3个关键模块，并得到6个预后相关基因，为研究结直肠癌的转移机制及预后、靶向治疗提供了一定的理论支持。

Abstract:

Objective: To explore the mechanisms of colorectal cancer (CRC) metastasis and find the potential biomarkers associated with metastasis and prognosis by analyzing the gene profiles with bioinformatics analysis. Methods: The gene profiles of GSE41568, GSE68468 were selected from the GEO database; R software with the limma package was used to analyze the data and screen the differentially expressed genes between primary tumors and metastatic tumors, and we got the common differentially expressed genes. ClusterProfiler package was used for the biological function enrichment analysis and visualization; STRING database and Cytoscape software were used to construct protein-protein interaction (PPI) network and identify the key modules; TCGA database was applied for the expression analysis and survival analysis of common differentially expressed genes. Results: We identified 108 common differentially expressed genes between primary tumors and metastatic tumors. Enrichment analysis indicated that the genes were enriched in complement and coagulation cascades pathways and biological processes, et al. We also found 3 key modules by PPI network analysis. The analysis of common differentially expressed genes based on TCGA data showed that there were significant differences in the expression of CLCA1, COLEC11, FCGBP, PDZD2, SERPINA1, and SPINK4 between the stage Ⅰ-Ⅱ and Ⅲ-Ⅳ of CRC (P<0.05) , with significant correlation with the prognosis of CRC (P<0.05). Conclusion: With bioinformatics analysis, a total of 108 common differentially expressed genes, 3 key modules and 6 genes associated with prognosis have been screened out, thereby providing the theoretical support for research on mechanisms of metastasis, prognosis prediction and targeted therapies of CRC.