Comparison of Effect of Non-Specific Filtering Methods and their Combinations on GeneChip Data Analysis: A Case for Affymetrix Porcine Genome Microarray

摘要

Microarray technology has been widely applied in the research area of animal genetics and breeding. In the earlier stage of microarray data analysis, the non-specific filtering is a common procedure which has been used increase the detection rate of differentially expressed genes. In this investigation, researchers use the data from Affymetrix GeneChip Porcine Genome Array to comparatively assess the effects of different non-specific filtering methods on the results of transcriptome analysis. The research results showed that the MAS5, ET and IQR filters could increase but the I/NI-calls method could decrease the number of differentially expressed genes. Here into, the IQR filter has the largest detection rate. Furthermore, the two-way combinations with IQR filter have the similar increasing effect on differentially expressed gene detection. It is concluded that not all non-specific filtering methods could increase the detection rate in microarray data analyses and the IQR filter could be considered as a preferred choice when improvement in the detection rate is needed on some occasions.