Hi,
I have 100 bigwig files and a bed file which has 80 features. I want to load all the bigwig files and bed annotation file in IGV and for each feature I want to output an image which has for each feature all the loaded 100 bigwig files. In total I should have 80 graphs(for 80 features). Also I want it to scroll and not a screen shot as I want to see all the bigwig files. I am writing a batch script as told in IGV but I cannot get all the bigwig files to display.
Any help how to do it is appreciated.
Thanks
•
written
by
Varun Gupta
You can do this with batch files in IGV. Here's a little wrapper script that automates some of it: https://github.com/chrisamiller/igvScreenshot
To be clear, I know you said no screenshots, but on a headless node, you can expand these to whatever height/size you want by tweaking params.
In the "screenshot" categorie, check this paper : Strategies for identification of somatic variants using the Ion Torrent deep targeted sequencing platform. They have "created an IGV batch script to load the samples in which each variant was called", it's pretty nice.
Recently I started using https://github.com/deeptools/pyGenomeTracks for a similar purpose. You can loop over all features or run in batch mode.
You have to spend some time configuring the input tracks file though.
Hi venu, I downloaded and installed the pyGenomeTracks but I cannot locate make_tracks_file command for making the config file. Where does this go. I am using linux machine
pip install pyGenomeTracks
It got installed here
/home/varun/.local/bin
Hi Venu,
I tried making plots by providing this command
./pyGenomeTracks --tracks tracks.ini --BED ~/test.bed -out fig_
The graph is showing too much upstream and downstream of my bed coordinates. Any way to fix that Also how can i display the name of the feature present in the bed file??
the Bioconductor Gviz package will also do what you're looking for (allowing for Genome Browser style snapshots in a scripted manner)
In these cases, I cannot help mentioning my own tool, ASCIIGenome. It doesn't help with the question of how to do it with IGV, of course, but I would suggest that for batch processing it might be a better alternative (see examples here).
Log in to answer this question.
