计算机应用与软件2017，Vol.34Issue(11)：247-251,5.DOI:10.3969/j.issn.1000-386x.2017.11.046
基于模拟退火优化双聚类的基因数据填补方法
GENE DATA IMPUTATION METHOD BASED ON SIMULATE ANNEALING OPTIMIZED BICLUSTERING
摘要
Abstract
Gene expression data are generated from the DNA microarray experiments of large data matrix,which can effectively extract the biological information.Due to the limitation of experimental conditions,gene expression data is often the presence of missing values and need to fill the missing data.The traditional missing data imputation method is based on the single feature of the gene expression data,and the correlation between the data matrix is not considered.The smaller the mean square residue of the dual clustering is,the higher the correlation between the data of the gene expression data is,and a method of missing data filling (bi-SA) based on simulated annealing optimized double clustering is proposed.This missing data imputation method used simulated annealing to optimize the optimal biclustering,and thus achieved the most effective imputation of missing data.Four groups of real gene expression data show that the bi-SA method has higher accuracy compared with other imputation methods.关键词
基因表达数据/缺失数据填补/模拟退火法/双聚类Key words
Gene expression data/Missing data imputation/Simulated annealing algorithm/Biclustering分类
信息技术与安全科学引用本文复制引用
朱娴,杨明,马卫,朱俊..基于模拟退火优化双聚类的基因数据填补方法[J].计算机应用与软件,2017,34(11):247-251,5.基金项目
江苏省高校自然科学研究项目(15KJB520017,16KJB520019). （15KJB520017,16KJB520019）
