基于MST的基因数据社团挖掘算法

Detecting communities of gene expression data based on MST

使用机器学习方法来分析生物信息学中一些复杂的基因表达数据是目前重要的研究领域之一。使用社团挖掘的方法对基因表达数据进行分类,社团内由类似的基因数据组成,研究和分析每个社团的结构和功能以及社团之间的关系,这对深刻认识诸多生物过程的本质有重要意义。将最小生成树的概念引入生物信息学中基因表达数据的社团挖掘分析中,设计了最小生成树来表示基因表达数据和基于此的社团挖掘算法,针对该算法提出一些目标函数,来判别基因表达数据社团挖掘算法的性能。最后,通过实验验证了该算法对于一些目标函数能够产生最优的社团划分,并且社团挖掘算法的性能良好。

With the development of the bioinformatics, how to analyzes complex genomics data using machine learning approach has become an important research field. Gene expression data are classified by community mining method, and the community is made up of the similar genetic data. It is very important significance for understanding the essence of biological processes that the structure and function of community and the relationship between the communities are researched. Minimum spanning tree is used in community detection gene expression data of molecular biology. Using minimum spanning tree to represent the gene expression data and detecting community method are designed. The performance of detecting community method is evaluated for some rule function. According to the results of the experiments, it is proved that optimum community can be obtained and the performance of the algorithm is good.