摘要
生物基因表达数据的双向聚类已成为近年来生物信息学的研究热点.生物基因表达数据双聚类问题通常需要同时最优化双聚类中基因表达行为的波动一致性以及双聚类的容量.基于单目标优化的双聚类算法难以很好地同时优化这两个目标.针对这个问题,本文采用了多目标微分进化算法来求解基因表达数据的双向聚类问题.算法在真实的基因表达数据集上测试,实验结果表明,本文所提算法具有更优的聚类效果.
Biclustering of the gene expression data has become a research focus in bioinformatics in recent years. Usually, the fluctuat consistency of gene expression activities and the volunm of biclusters are to be optimized at the same time when we solve the biclustering problem of gene expression data. However, it is difficult for a single-objective evolutionary biclustering algorithm to optimize these two objects at the same time. To address this issue, this paper employs a multi-objective differential evolution algorithm to solve the biclustedng problem. The algorithm proposed in this paper was tested in real gene expression datasets. Experimental results show that the proposed algorithm has better clustering effect.
出处
《小型微型计算机系统》
CSCD
北大核心
2010年第10期1997-2001,共5页
Journal of Chinese Computer Systems
基金
福建省自然科学基金项目(2009J01283)资助
关键词
双聚类
基因表达数据
多目标优化
微分进化算法
biclustering
gene expression data
multi-objective optimization
differential evolution algorithm
