a scaling normalization method for differential expression analysis of rna-seq data一个比例微分表达式rna-seq分析数据的归一化法.pdfVIP

Robinson and Oshlack Genome Biology 2010, 11:R25 /2010/11/3/R25 METHOD Open Access A scaling normalization method for differential expression analysis of RNA-seq data Mark D Robinson1,2*, Alicia Oshlack1* Abstract The fine detail provided by sequencing-based transcriptome surveys suggests that RNA-seq is likely to become the platform of choice for interrogating steady state RNA. In order to discover biologically important changes in expression, we show that normalization continues to be an essential step in the analysis. We outline a simple and effective method for performing normalization and show dramatically improved results for inferring differential expression in simulated and publicly available data sets. Background sophisticated normalization of data sets’ [5]. We demon- The transcriptional architecture is a complex and strate here that the reality of RNA-seq data analysis is dynamic aspect of a cell’s function. Next generation not this simple; normalization is often still an important sequencing of steady state RNA (RNA-seq) gives unpre- consideration. cedented detail about the RNA landscape within a cell. Current RNA-seq analysis methods typically standar- Not only can expression levels of genes be interrogated dize data between samples by scaling the number of without specific prior knowledge, but comparisons of reads in a given lane or library to a common value expression levels between genes within a sample can be across all sequenced libraries in the experiment. For made. It has also been demonstrated that splicing var- example, several authors have modeled the o