Wald-Log Test – most powerful statistical method of detecting differentially expressed genes in RNA-Seq data sets

To detect differentially expressed genes under two conditions, statistical methods such as Poisson distributions are often used. However, to accurately detect differential expression of gene with low expression levels, more powerful statistical methods are desirable. In statistical literature, several methods have been proposed to compare two Poisson means (rates).

Through simulation study and real data analysis, the authors find that the Wald test with the data being log transformed is more powerful than other methods, including the likelihood ratio test, the variance stabilizing transformation test, the conditional exact test and the Fisher exact test.

When the count data in RNA-seq can be reasonably modelled as Poisson distribution, the Wald-Log test is more powerful and should be used to detect the differentially expressed genes.

  • Chen et al. (2011) Statistical methods on detecting differentially expressed genes for RNA-seq data. BMC Systems Biology 5(Suppl 3), S1. [article]