Differential Expression Analysis Based on Expression Data of Multiple Platforms

Abstract

Differential expression analysis is one of the basic objectives of the study of the transcriptional group, which plays an important role in revealing the function and regulation of genes, as well as the fluctuation of selective shear. Many researches have adopted many measuring platforms to get more accurate results. Research shows that fusion of multi-platform expression data can improve the accuracy and reliability of differential expression analysis. Most of the existing differential detection studies of fusion multi-platform expression data are mainly fusion of various types of gene chip expression data, and less RNA-Seq expression data are considered. Moreover, the existing methods ignore many useful information, such as measurement errors and the volatility produced by repeated experiments. In view of the problems existing in the existing methods, this paper proposes a differential detection model mpDE (multi-platform Differential Expression model), which combines multi-platform transcriptome data to find differentially expressed genes and isomers. The model integrates the technical measurement errors of the expression data and the expression level of the different experimental platforms into the model, and takes into account the fluctuation of the same platform in different conditions of biological repetition or technical repetition, thus improving the accuracy of the difference detection. In this paper, mpDE is applied to two human multi-platform expression data sets for differential expression detection. It involves the traditional 3’ chip, exon arrays, HTA2.0 chip, and RNA-Seq four common transcriptional level measurement platform for Affymetrix.