[BioC] question about Gviz

Hahne, Florian florian.hahne at novartis.com
Thu Aug 7 11:16:07 CEST 2014


Hi Yu,
I am not quite sure what you mean by ³increase the resolution². Your
display or any graphics device that you chose for your Gviz plot has a
finite number of pixels available for plotting. Now if you selected a
device with dimensions 500x500 pixels, the minimum width that you can
distinguish on that plot will be roughly width/500. In genomic coordinates
that is typically not much, and it may well be that your 5kb bins have
exactly the same size as your 5bp bins.
The simple solution is to pick a wider device, e.g, do png(width=2000,
height=500) instead of png(width=500, height=500). That being said, there
are plotting parameters that control the minimum width of a range
(min.width and min.distance, you may want to check the documentation for
the RangeTrack class for details).
Florian

From:  Yu Zhou <y6zhou at ucsd.edu>
Date:  Thursday 7 August 2014 07:30
To:  Florian Hahne <florian.hahne at novartis.com>
Subject:  question about Gviz


Dear Florian,

Thanks a lot firstly for developing Gviz! I'm using it a lot and refer it
in my recent manuscript.

One question is how to improve the resolution for histogram figure? When
making the figure small,  50 bp interval looks like the same width as 5000
bp or even longer intervals. Any suggestion will be appreciated. Thanks
again!

Best regards,
Yu



-------------------------------------------------------
Yu Zhou, Ph.D.
Postdoc in Fu Lab
Department of Cellular and Molecular Medicine
School of Medicine
University of California, San Diego
9500 Gilman Dr, La Jolla, CA 92093-0651
Lab Phone: 858-534-4926



More information about the Bioconductor mailing list