期刊
JOURNAL OF STATISTICAL PLANNING AND INFERENCE
卷 138, 期 2, 页码 500-515出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.jspi.2007.06.019
关键词
gene expression data; biplot; supplementary data; principal component analysis; factor analysis; correspondence analysis; multidimensional scaling
DNA microarray experiments result in enormous amount of data, which need careful interpretation. Biplot approaches show simultaneous display of genes and samples in low-dimensional graphs and thus can be used to represent the relationships between genes and samples. There are several different types of biplots, and these methods need to be evaluated because each plot provides different result. In this paper, we review several variants of biplot methods such as principal component analysis biplot. factor analysis biplot, multidimensional scaling biplot and correspondence analysis biplot. We investigate the properties of these methods and compare their performances by analyzing various types of well-known gene expression data. We also suggest the supplementary data method as a tool for (i) classifying the previously unknown sample/gene to existing class, (ii) analyzing mixture data and (iii) presenting illustrative variables, etc. The usefulness of this approach for interpreting microarray data is demonstrated. (C) 2007 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据