摘要
微阵列又称为基因芯片,基因芯片数据的特点是基因数多而样本数少,也就是变量数(或基因数)p远高于样本数n,表现出"大p,小n"。这对分析过程带来一定困难。实验表明,主成分分析方法能有效地解决这一问题。通过特征基因提取,提高基因芯片数据分析的准确性。
Microarray is called gene chips, characteristics of gene chip data is gene large and the size of sample is small, that is the number of variables (or genes) 'p' is much higher than the samples 'n', showed "the big 'p', little 'n' ". It is difficult to analyze the process. It shows that the method of principal component analysis can solve the problem effectively. The accuracy of gene chip data analysis can improve by extracted feature gene.