Estimating differential expression with DESeq2
The DEseq2 package is a popular tool for performing differential analysis of count data, so it is ideal for expression analysis of RNA-Seq data in R and other count data such as ChIPSeq.
DEseq2 performs normalization using a method called variance stabilizing transformation (VST), which is a type of transformation that aims to stabilize the variance of the data across the range of counts. This is in contrast to other normalization methods that aim to bring the mean of the data to a specific value, such as the mean of all the samples or the median of all the samples. The VST method is effective at reducing the variance of the data estimating with and improving the statistical power of differential expression analyses. This allows us to focus on improving gene ranking in results tables.
DEseq2 uses a negative binomial model to fit the count data and estimate the dispersion parameter. This model is commonly used for RNA-Seq data because...
