first bases of each chromosomes missing when loading genome file

[Pierre Grognet]

Hi,

I have a multifasta file of my genome sequence where chromosomes are named Chr1, Chr2, etc.

When I load this genome with IGV, the first base of chromosome 1 is missing, the 2 first bases of chromosome 2 are missing, 3 bases for chromosome 3, etc.

So when I load an annotation file everything is shifted.

This problem occurs with IGV 2.16, 2.17, 2.18.4 on Windows but not with IGV 2.18 for Linux (Ubuntu) with the exact same files.

Thanks for your reply, I'll report this.

For example, the chromosome 2 starts with this sequence:

>Chr2
cttaactttatatttttta...

But here is what IGV is displaying

Does anyone have any idea on how to solve that problem?

Thanks!

[igv-help]

Hi,  I can't reproduce this issue.   Can you zip and post a fasta exhibiting the problem (including the index file).   If its too large to post email igv-...@broadinstitute.org and we can give you a dropbox link for upload.

[James Robinson]

BTW just a guess, but if the files were indexed on a Linux or Mac machine,  then transferred to  Windows in a way that changed line endings to Windows convention, that could cause this issue.  In doing that (changing line ending conventions) you will have invalidated the index by 1 byte per line.   Again this is just a guess.

Compare the exact sizes in bytes of the files on the Windows and Linux machines.   

--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/HKE4ISRlKxU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/igv-help/44b72a08-0a02-4437-af25-d4485957ebe0n%40googlegroups.com.

[Pierre Grognet]

Hi,

Thanks for your reply.

I already tried to play around with the line endings but it didn't make things better.

After reading your message, I deleted the files from my windows computer and copied them again from my Linux one and the problem was gone.

In these files, the line endings follow Linux convention, except for the fasta headers that end in a Windows convention:

This way it works well

Best

Pierre Grognet

To view this discussion visit https://groups.google.com/d/msgid/igv-help/CACOP%2BptGpKSDkOjfK_img0bWdn_4fTCOqnM02rBOcqWF%3D8-GDQ%40mail.gmail.com.