论文部分内容阅读
生命科学已进入以功能基因组研究为主的后基因组时代,基因微阵列技术是功能基因组学研究领域最常用的手段,可同时获得大量基因的表达谱数据资料。发现差异表达基因是微阵列实验研究的主要目的之一,本文针对微阵列实验中差异表达基因常用统计分析方法进行综述。多序列两样本比较时,t检验法是最简单的检验差异表达基因的统计分析方法。多序列多组比较时,统计推断可采用方差分析,其中混合效应方差分析(方差分量模型)是含有多个误差来源的多因素微阵列实验有效统计分析方法。其他线性和非线性混合效应模型用于基因表达微阵列数据的统计学分析有待进一步研究。
Life science has entered the era of post-genomics with functional genomics research. Gene microarray technology is the most commonly used method in the field of functional genomics, and can obtain the expression profile data of a large number of genes at the same time. The discovery of differentially expressed genes is one of the main purposes of microarray experiments. In this paper, the commonly used statistical methods for the differential expression of genes in microarray experiments are reviewed. When comparing multiple sequences with two samples, the t-test is the simplest statistical analysis to test differentially expressed genes. Variance analysis can be used for statistical inference when multiple sequences and multiple groups are compared. The analysis of variance of variance (variance component model) is a multi-factor microarray experimental statistical analysis method with multiple sources of error. Other linear and nonlinear mixed-effects models for statistical analysis of gene expression microarray data are for further study.