Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

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

Fig. 3

(Top) Example of a ChIP-seq coverage profile annotated by an expert biologist. The labels represented by colored rectangles indicate the absence or presence of a peak, here characterized by its start and its end . (Bottom) The model with 1 peak in its output segmentation has an associated error of 2 (\(2\times\) False Negative ). The model with 3 peaks has an associated error of 1 (\(1\times\) False Positive ). The model with 2 peaks is a good model for which all the labels optimized on this coverage profile are correct

Back to article page