利用文本挖掘技术和生物信息学方法研究鹿茸生物学功效

Explore the biological efficacy of Pilose Antler using text mining technology and bioinformatics method

目的利用文献数据文本挖掘技术结合生物信息学分析方法 ,从基因表达谱角度探索鹿茸的生物学功效。方法首先对中国知网、万方数据库和维普数据库三个常用的中文数据库中关于鹿茸的基因表达调控作用的相关文献进行挖掘,提取基因表达数据,进行数据清洗、标准化。其次,通过Cytoscape制作蛋白相互作用网络,再利用自主研发的"疾病-基因-化合物"分析平台进行关联分析,得到鹿茸的适应证和禁忌证。使用DAVID工具对鹿茸基因表达数据进行KEGG通路功能富集分析,筛选出相应的生物学通路及靶点。最后,通过查阅鹿茸药理学作用相关古籍、文献验证本方法的准确性。结果本文共挖掘到鹿茸相关调控基因30个,其中上调基因21个,下调基因9个;制作出鹿茸相关调控基因蛋白互作网络;关联分析找到十几种鹿茸的适应证和禁忌证;KEGG通路富集分析找到鹿茸在抗肿瘤、抗感染、提高免疫力三个功效方面的作用通路及靶点。结论文本挖掘结合生物信息学方法是一种研究中药药理机制及功效的新方法。

Objective To explore the biological efficacy of Pilose Antler using the method of literature data mining combined with bioinformatics analysis method from the view of gene expression profiles. Methods Firstly, text mining technology was used to explore literatures on gene expression regulation of Pilose Antler from CNKI, Wanfang and CQVIP Data. Gene expression data were extracted, duplicates were removed, and then the gene names were normalized.Secondly, Cytoscape was used to make the protein-protein interaction network. Then the association analysis was used to get the indications and contraindications of disease treatment by the ＂disease-gene-compound＂ analysis platform developed by our team. Next, DAVID online analysis tools was used to enrich and analyze the KEGG pathway functions and screen out the corresponding biological pathways and targets. At last, the accuracy of this method was tested and verified by referring to the ancient books and literature on the pharmacological effects of Pilose Antler. Results In this paper, 30 antler-related regulatory genes were extracted, of which 21 were up-regulated and 9 were down-regulated;the protein-protein interaction network of antler-related genes was produced; the association analysis found more than a dozen species indications and contraindications of Pilose Antler; KEGG pathway enrichment analysis found the pathways and targets of Pilose Antler in three functions of anti-tumor, anti infection and enhancing immunity. The feasibility of this method was verified by consulting pharmacological literature. Conclusion Text mining combined with bioinformatics analysis is a new method to explore pharmacological mechanisms and efficiencies of traditional Chinese medicine.