This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Convenient way to visualise a collection of genomic locations

Is anyone aware of a tool that would allow me to:

  1. upload a list of genomic locations, say a BED file.
  2. go through all of these locations one by one (i.e. by clicking next/next/next or similar) and
  3. visualise my choice of tracks at these locations?

I know genome browser does almost everything, but not the step #2, or am I mistaken?

visualisation chip-seq genome-browser
• 3,197 views • link
updated by Ram Ram • written by Saulius Lukauskas Saulius Lukauskas

So pretty much the consensus of the answers seems to be: "No. You need to write a less or more complicated script to do this." with plenty of suggestions on how to do this below.

• 0 views • link
updated by Ram Ram • written by Saulius Lukauskas Saulius Lukauskas

Behind every convenient technology, there are usually people sweating the details on your behalf, yes.

• 0 views • link
written by Alex Reynolds Alex Reynolds

What exactly would have merited a "yes, there is a convenient way" in your opinion?

IMHO, running bedTools once and subsequently loading the script beats having to click "next" for x number of times, but I may have misunderstood your intention.

• 0 views • link
written by Friederike Friederike

3 answers

Given your (UCSC-formatted) BED file, you could use the ucsc-fetch script to make a folder of images containing UCSC Genome Browser shots. The track selection is based on a tracks file that describes what is shown and how it is shown (full, dense, etc.).

Once you have that folder of images, you could use a tool like SimpleViewer to make a gallery. You might use a Perl or Python script that makes the required XML-formatted gallery file (this describes how the gallery is constructed and what images go into it), copies images to the images subfolder, and uses ImageMagick to make thumbnails to copy to the thumbs subfolder.

Images in the resulting SimpleViewer gallery can be opened in a web browser and you can navigate through them with arrow keys or mouse-click events. It's a pretty slick bit of gear.

I use a variation of the above recipe to make plot galleries for internal use in our lab. I use an internally-written equivalent of ucsc-fetch that works with internal UCSC Genome Browser instances, and hooks into SimpleViewer Pro for some extra browser functionality - for example, to offer a link to internal Genome Browser instances at the image's equivalent BED coordinate range and browser session ID, to load specific tracks.

It works pretty great. For various reasons, I'm not sure we can release our internal ucsc-fetch equivalent to the public, but here's the Perl script I use to make a SimpleViewer-based gallery XML file and other resources from an input folder of images. The use of this Gist by others would need some local customization, but perhaps it can help save some time:

You'll need to install ImageMagick, have a folder containing SimpleViewer or SimpleViewer Pro resources available, and a Perl library installed called XML::Writer to handle making the gallery XML file. But with an hour or two of coding and tweaking, the arrangement I describe above can be pretty easily repeated by others.

• 0 views • link
updated by Ram Ram • written by Alex Reynolds Alex Reynolds

The IGV Genome Browser lets you run batch scripts (open the session with all the tracks the way you want them, go to "Tools" --> "Run batch script"). The batch script could simply contain the genomic regions you're interested in and the command to take snapshots, more explanations here. Handily, BEDtools lets you make an IGV batch script directly from your original BED file: page 82 of this pdf

So, you would:

  1. Run bedtools bedtoigv -i your_regions.bed > your_batch_script.txt
  2. Open IGV
  3. Load all the tracks into IGV
  4. Go to "Tools" --> "run batch script" --> select your_batch_script.txt
  5. Sit back and relax while IGV is taking all the snapshots at the regions you specified
• 0 views • link
updated by Ram Ram • written by Friederike Friederike
  • Upload a BED file of your regions of interest in the UCSC genome browseras a custom track.
  • Run the following awk script in your BED to get the hyperlinks to the regions (e.g with genome=hg38:
cat your.bed | \
  sed -e 's/^chr//' -e 's/^/chr/' | \
  awk 'BEGIN{printf("<html><body>\n");} {printf("<a href=\"http://genome.ucsc.edu/cgi-bin/hgTracks?db=hg38&position=%s%3A%s-%s\" target=\"_blank\">%s:%s-%s</a>\n",$1,$2,$3,$1,$2,$3);} END{printf("</body></html>\n");}' > explore.html
• 0 views • link
updated by Ram Ram • written by Pierre Lindenbaum Pierre Lindenbaum

Log in to answer this question.

More posts like this