Home
pARP (Processing Abnormal ReadPairs) is basically nothing more than a genome browser that allows for showing features that are linked. Rather than aiming at presentation such as Circos, pARP is interactive and meant to be used for browsing and discovery.
See here for a short screencast.
One of the design decisions in pARP is that it shows context. Even if you’re not focussed into a region, any links from your focus regions that go there are still displayed. This means that the whole genome is always visible, albeit at different resolutions.
The display starts with the whole genome shown at a given resolution. By clicking on the circle you create a new focal point that takes up a fourth of the circle with a resolution that is 100 times higher. The result is different slices each with a different resolution. The slice under the mouse is the active one.
At the right of the display is an overview of all slices with their relative resolutions.
The interaction is not completely worked out yet and much of it is done with the keyboard at the moment:
- r: *R*eset the slices.
- b: Go *b*ack. A history of 20 steps is kept in memory.
- c: *C*ollapse current slice. This compresses it to 5 pixels.
- UP: Zoom in for current slice.
- DOWN: Zoom out for current slice.
- LEFT: Pan current slice to 5’ end. This changes the contents of the slice; the pixel boundaries are not changed.
- RIGHT: Pan current slice to 3’ end.
Mouse interaction:
- Click on circle to create new focus (see above).
- Drag blue dot at border between slices to change slice boundaries.
This tool is not ready for production use yet. However, you can already play around with it. There will still be quite some bugs, including several edge cases that are not captured yet. Use at your own risk.
There are applications for Mac, Windows and linux:
- Mac: pARP.app
- Windows: There’s a parp.exe program in the pARP.app directory.
- linux: There’s a parp command in the pARP.app directory.
Data has to be stored in a subdirectory of pARP.app/lib/data/. The directory name should be set in pARP.app/lib/data/config.yml. Filenames are hardcoded at the moment; files that have to be in the data directory:
- copy_number.txt
- genes.txt
- meta_data.tsv
- read_pairs.txt
- segdups.txt
More information about the configuration and data file formats can be found here: FileFormat.
See the Developer page.
Many thanks go to Jeremy Ashkenas for writing the ruby-processing library and improving (or at least: “changing”) it so that it can do what I needed it to do. Also thanks to Klaudia, Erin, Jon, Nelo and Chris for their ideas.