摘要
在区分肿瘤样本与正常样本的过程中,维数过多的基因表达数据会影响最终的分类结果.针对这一情况,在去除冗余基因的过程中,利用相关系数矩阵M构建强相关树,得到一种去除冗余基因的强相关树(Strong Correlation Tree,SCT)算法.实验结果表明,SCT算法能够去除更多的冗余基因,使最终的分类结果更加准确.
In the process of identifying the tumor and normal samples, dimension of gene expression data will affect the final classification result. In view of the situation, in the process of removing redundant genes,we use correlation coefficient matrix M to construct a strong correlation tree, so that we can get the Strong Correlation Tree algorithm (SCT). Comparative experimental results show that the proposed SCT method can remove more redundant genes, so as to make the final result more accurate.
出处
《内蒙古师范大学学报(自然科学汉文版)》
CAS
北大核心
2015年第6期757-760,共4页
Journal of Inner Mongolia Normal University(Natural Science Edition)
基金
内蒙古自然科学基金资助项目(2013MS0116)
关键词
相关系数
相关系数矩阵
强相关树
基因表达数据
correlation coefficient
correlation coefficient matrix
strong correlation tree
gene expression data