Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

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

Fig. 2

(Top) Segmentation of a coverage profile containing one peak using the unconstrained model. The location of the changepoints on the chromosome are shown by red dotted lines . The mean of the segments are shown in blue . According to this segmentation there are two alternative starts and two alternative ends of the peak, i.e. four alternative variants of the same peak formed by the regions: [Start1:End1], [Start1:End2], [Start2:End1] and [Start2:End2]. (Bottom) Three different rules are proposed to interpret the segmentation as peaks. Thinnest peak: the resulting peak is defined by the region [Start2:End1]. Largest peak: the resulting peak is defined by the region [Start1:End2]. Max jump: the resulting peak is defined by the region [Start1:End1]

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us