error loading genome file - deposit with full genome file avilable for download ?

802 views
Skip to first unread message

Jan Attig

unread,
Mar 30, 2016, 8:30:13 AM3/30/16
to igv-help



I'm running Mac OSX 10.10.5 (Yosemite), and IGV_2.3.71 with the bundled Mac App. I can not load genome files

error message
"The genome file [/Users/attigj/igv/genomes/hg38.genome] could not be read. Would you like to remove the selected entry?"

[log file attached].


I made 2 attempts to solve the issue.

(a) I downloaded hg38.genome from  http://igvdata.broadinstitute.org/genomes/hg38.genome, in case my file is corrupted. error remains

(b) From similar error messages in the forum, I started to think it's due to the proxy from our intranet to the internet. I fiddled with the proxy but was unable to set it so that IGV does connect.

b[i] I noticed a bug here. setting the proxy in the Preferences,  closing the preference window, back opening the preferences/proxy panel again. IGV does not remember the authenticication but does keep the box active. Is that meant to be like that?


Now my question: I know I can built the genome file myself with fasta/fai + .gff   Now my question is,
[i] will this enable me to run IGV even in the absence of internet connection ?
[ii] The hg38.genome file (see above) is a binary. I assume it dynamically links IGV to the human genome sequence on the broadinstitute server. Can I download the 'full' IGV-compatible reference file from a repository? Is that what the self-built .genome file will essentially achieve?


Thank you for your time & efforts

 Jan



igv.log

James Robinson

unread,
Mar 30, 2016, 2:56:01 PM3/30/16
to igv-...@googlegroups.com
Hi,

If you are having issues with interent connections your best option might be to build the genome yourself.    The reference fasta came from the UCSC downloads page, which you can download from here: http://hgdownload.cse.ucsc.edu/goldenPath/hg38/bigZips/hg38.fa.gz

After downloading,  gunzip it and load it directly from the Genomes > Load from File menu.   The first time you load it it will index it automatically, this will take a few minutes.   Its not neccessary to create a “.genome” file,  you can just load annotations from the file menu,  but if you wish to the annotation (gene) file we use is also downloaded from UCSC, specifically


Save this file locally and gunzip it before using for a “.genome” file.   Again, the .genome file is optional,  a fasta is enough to define the reference.

Hope this is helpful.   

Jim

--

---
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+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/3091d5a8-1639-49fa-97e6-3b4699ee32e8%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
<igv.log>

Jan Attig

unread,
Mar 31, 2016, 7:28:36 AM3/31/16
to igv-help

Hi Jim,

 thank you. I've built it with the links you've highlighted, and it's running nicely.

Jan


Reply all
Reply to author
Forward
0 new messages