糖尿病基因药物网络模型构建与分析

Construction and analysis of diabetes mellitus genes- drugs network model

目的:深度挖掘海量生物医学文献中糖尿病相关基因药物之间的隐含关联。方法:以糖尿病为研究起点,先对在Pub Med数据库中检索到的文献进行规范化数据清洗,然后提出一种基于网络的计算框架,并依据"ABC理论"来构建糖尿病的基因与基因、药物与药物、基因与药物这3种不同关联,最后使用R语言实现关联网络模型。结果:得到糖尿病相关基因有23种,它们之间存在43种不同关联;得到糖尿病相关药物34种药物,它们之间存在64种关联;得到糖尿病相关基因药物网络中含有10种基因和34种药物,它们之间存在40种关联。结论:糖尿病的基因药物之间存在大量程度不同的关联,为糖尿病个性化医疗提供了研究依据和研究思路。

Objective： To have a deep mining for the hidden connections between diabetes mellitus interrelated genes and drugs,in the vast amount of biomedical literature. Methods： Using diabetes mellitus as the research start,the literature in Pub Med database was retrieved and downloaded first,and then standardized with data cleaning. Secondly,according to ＂ ABC- principle＂,a kind of computing framework based on network was proposed to construct 3 kinds of different diabetes mellitus relevance（ genes- genes,drugs- drugs and genes- drugs）. Finally,the R language was used to implement the associated network model. Results： The results of the experiment were 23 diabetes mellitus associated genes and 43 associations between them; and 34 drugs and 64 associations between them; there were 10 genes and 34 drugs in the genes- drugs network,with 40 association between them. Conclusion： There are a large number of different associations between the gene and drugs of diabetes mellitus,which provide foundation and research ideas for diabetes- mellitus- personalized medicine.