基于逐步回归分析的基因表达缺失值估计

A Method to Estimating Missing Values for Microarray Expression Data by Stepwise Regression Analysis

在基因芯片实验中,数据缺失客观存在,并且在一定程度上会影响芯片数据后续分析结果的准确性。在不增加实验次数的情况下,缺失值估计是降低缺失数据对后续分析影响的有效方法。针对基因表达数据的特点,提出了基于逐步回归分析方法的基因表达缺失值估计算法。实验结果表明,新的估计算法具有较传统缺失值估计算法更好的稳定性和估计准确度。

In microarray experiments,the missing value does exist and somewhat affect the stability and precision of the expression data analysis.Compared with increasing experiments,missing value estimating is preferred in reducing the influence of missing values on the post-processing.Considering the additive noise and high dimension in the expression dataset,a new method based on Stepwise Regression Analysis（SR） is presented.Experimental results show that the novel method has better performance than the existing methods that have been employed.