基于逻辑运算的离散人工蜂群优化双聚类算法

Discrete Artificial Bee Colony Optimization Biclustering Algorithm Based on Logic Operation

基因表达数据是由DNA微阵列实验产生的大规模数据矩阵,双聚类算法是挖掘数据矩阵中具有较高相关性的子矩阵,能有效地提取生物学信息。针对当前多目标双聚类优化算法易于陷入早熟和局部最优解等问题,论文提出了基于逻辑运算的离散人工蜂群优化双聚类算法(LOABCB算法),一方面引入人工蜂群算法增强双聚类的全局寻优能力,另一方面通过逻辑运算邻域搜索策略寻找最优双聚类,提高搜索效率。采用基因表达数据的酵母细胞数据集进行实验,结果表明论文算法能够获得实验效果优的具有生物意义的双聚类。

Gene expression data is generated from the DNA microarray experiments of large data matrix,which can effectively extract the biological information.Aiming at the multi-objective optimization biclustering algorithm is easy to fall into the problem of premature convergence and local optimal solution,this paper proposes a discrete artificial bee colony optimization biclustering algo⁃rithm based on logical operation(LOABCB algorithm),on the one hand,artificial bee colony algorithm is introduced to enhance the global searching ability of biclustering,on the other hand through the logical operation neighborhood search strategy to find the optimal bicluster,the search efficiency is improved.Evaluating the performance of the algorithm is a famous gene expression dataset of yeast cells based on data sets.The experimental results show that the algorithm can obtain biclusters with experimental effect and biological significance.