互信息关联模型在基因调控网络构建中的应用  

The mutual information correlation model's application in construction of gene regulatory net-work

作　　者：王锐[1] 张献志[1] 曾丽华[2] 陈活良[1] 张洁琼[1] 

机构地区：[1]第四军医大学卫生勤务学教研室,陕西西安710032 [2]第四军医大学放射医学教研室,陕西西安710032

出　　处：《西部医学》2018年第1期8-11,共4页Medical Journal of West China

基　　金：国家自然科学基金(51677189);陕西省自然科学基金(2014JM4090)

摘　　要：目的基因的表达受其它基因的影响较大,基因之间相互影响、相互制约的关系构成了复杂的基因表达调控网络。本研究基于基因表达的时空特性,寻求一种新的描述基因表达调控网络的方法。方法根据信息熵确定属性权重,对各时间点进行加权聚类;根据互信息相关理论,确定相关系数,分析基因之间的调控关系。结果得到各基因之间的调控矩阵,进一步进行图像可视化。结论利用信息熵理论得到信息熵相关系数后,可以根据两两基因之间信息熵相关系数的大小确定基因之间的调控关系,从而得到基因之间的生物学关系。该方法计算简便,结果更为合理,可以从整体上了解生物细胞的生长过程,把握基因之间的相互关系,特别是对于发现基因的新功能有较好的效果,并为进一步的生物学实验提供了相关依据。Objective The expression of the gene is affected by other genes. The complex relationships formed the so-called gene regulatory networks for genes. Based on time-space characteristics of gene expression, we searched a new description of gene regulatory net-work of genetic expression. Methods We try to use entropy to make sure weights, and weight clustering to several times. According to coefficient of entroy correlation, we described the regula- tive relation between genes. Results Getting the controlling matrix between genes, we could visualize the image. Conclu- sion After get the information entropy correlation coefficient by the use of information entropy theory, we can confirm regulatory relationships between two genes according to the size of the information entropy correlation coefficient and get the biological relationship between genes. The method is simple and result is more reasonable. We can understand the bi- ological cell growth process on the whole, to grasp the relationship between genes, especially to be effectively to find new genes function, and which can provide a basis for further biological experiments.

关 键 词：基因调控网络 互信息 信息熵 属性权重 

分 类 号：Q-33[生物学] TN911.72[电子电信—通信与信息系统]