by a number of factors including the degrees of freedom, the estimated mean counts μij, and the gene’s dispersion estimate αi. The prior influences the MAP estimate when the density of the likelihood and the prior are multiplied to calculate the posterior. Genes with low estimated mean values μij or high dispersion estimates αi have flatter profile likelihoods, as do datasets with few residual degrees of freedom, and therefore in these cases the zero-centered prior pulls the MAP estimate from a high-uncertainty MLE closer toward zero.
