Slow Viewing Speed for Files with Large Distance Between Paired Reads

[Dario Strbenac]

I have mapped some RNA-seq files allowing for large distances (100 million) between paired reads. It takes about 10 minutes to load the reads over a gene, even if there are very few of them. Moving the view even a little horizontally causes another many minutes wait before IGV shows the reads. This is much slower than when I used the default parameters of STAR previously where the view loads in a couple of seconds. Can this task be made quicker? Also, if I right click on a blue horizontal line that joins two reads and then left click on Go to mate, it just switches to another region with lots of horizontal lines that doesn't seem to contain either of the reads I am interested in. I am using IGV 2.3.81.

[James Robinson]

Dario, could you open a git issue?

On Thu, Sep 22, 2016 at 2:00 AM, Dario Strbenac <dario....@gmail.com> wrote:

I have mapped some RNA-seq files allowing for large distances (100 million) between paired reads. It takes about 10 minutes to load the reads over a gene, even if there are very few of them. Moving the view even a little horizontally causes another many minutes wait before IGV shows the reads. This is much slower than when I used the default parameters of STAR previously where the view loads in a couple of seconds. Can this task be made quicker? Also, if I right click on a blue horizontal line that joins two reads and then left click on Go to mate, it just switches to another region with lots of horizontal lines that doesn't seem to contain either of the reads I am interested in. I am using IGV 2.3.81.

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/366b9df7-ea3f-4383-9a6d-0007beca18cc%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[James Robinson]

Also include a sample file that reproduces the problem, if possible.  

There are optimizations in the snapshot version that might improve the situation you describe.