[Bioc-sig-seq] ChIP-seq, peak shape

Ivan Gregoretti ivangreg at gmail.com
Wed Apr 21 17:26:36 CEST 2010


Hello everybody,

Is there any Bioconductor tool designed to analyse/visualise the shape
OF A SET of ChIP-seq peaks?


Now the same question but with more words:

It is simple to visualise the shape of a single peak by computing the
coverage along the chromosome. The human eye can immediately identify
a peak's centre and its tails. Thus, one can tell peak width, peak
sharpness and, if it exists, bi-modality. A violin graph for instance
--see library(lattice)-- is a one step solution for its descriptive
statistics

A more complex problem is to try to compute a consensus shape out of
30000 ChIP peaks. I reckon that the peaks should be somehow centred
and then the tags consolidated relative to each centre. A violin graph
of the consolidated set would then finish up the task.

I wonder if a Bioconductor tool for this purpose already exists or I
will have to code it myself.

Thank you,

Ivan

Ivan Gregoretti, PhD
National Institute of Diabetes and Digestive and Kidney Diseases
National Institutes of Health



More information about the Bioc-sig-sequencing mailing list