Fig. 4
From: Inference of genomic landscapes using ordered Hidden Markov Models with emission densities (oHMMed)
As the final part of the summarised diagnostics for oHMMed with gamma-poisson emission densities and three hidden states as applied to the protein coding genes in humans, we present the above rootogram: The bars represent the observed frequency of counts (square root transformed), and they have been shifted so that the top of each bar aligns with the (smoothed) distribution inferred by oHMMed. Deviations can therefore be assessed by checking the distance of the lower end of each bar to the x-axis