edgeR estimates the genewise dispersions by conditional maximum likelihood, conditioning on the total count for that gene (Smyth and Verbyla, 1996). An empirical Bayes procedure is used to shrink the dispersions towards a consensus value, effectively borrowing information between genes (Robinson and Smyth, 2007). Finally, differential expression is assessed for each gene using an exact test analogous to Fisher's exact test, but adapted for overdispersed data (Robinson and Smyth, 2008).
