Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: Increased peak detection accuracy in over-dispersed ChIP-seq data with supervised segmentation models

Fig. 4

The over-dispersion exhibited by count data under a Poisson noise model can be effectively reduced using a negative binomial or a Gaussian transformed noise model. The red indicator line stands for the equality of the theoretical and empirical variances. (Left) Observations above this line stand for over-dispersed count data. Observations under this line stand for under-dispersed count data. (Right) A zoom on the Left distributions

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us