Question: Built-in indexes not working in Local Galaxy Instance
3
gravatar for samuel.tremblay-belzile
3.8 years ago by
Canada

Hi,

I've just finished setting up a local Galaxy instance and everything works fine except for built-in reference genomes both with the "Map with BWA for Illumina" and "Map with Bowtie for Illumina" tools. I am able to use the same reference genomes without any problem if I set them up as files in the history. I first tried to set up the .loc and fasta files manually as instructed on this page:

https://wiki.galaxyproject.org/Admin/DataPreparation

When using the Bowtie tool, I can't see any reference genomes in the drop down menu for built-in indexes. When using the BWA tool, I see the genomes, but I get the following error right as the task starts:

The alignment failed.
Error aligning sequence. [bwa_aln] 17bp reads: max_diff = 2
[bwa_aln] 38bp reads: max_diff = 3
[bwa_aln] 64bp reads: max_diff = 4
[bwa_aln] 93bp reads: max_diff = 5
[bwa_aln] 124bp reads: max_diff = 6
[bwa_aln] 157bp reads: max_diff = 7
[bwa_aln] 190bp reads: max_diff = 8
[bwa_aln] 225bp reads: max_diff = 9
[bwt_restore_bwt] fail to open file '/galaxy/data/Arabidopsis_thaliana_TAIR10/bwa_index/Arabidopsis_thaliana_TAIR10.fa.bwt'. Abort!
Aborted

I thought maybe I had made a mistake in setting up the genomes, so I used rsync to copy all the .loc files and the Arabidopsis genome from the public Galaxy server, but I get exactly the same error. This is strange, considering this alignment has been completed successfully on the public server in the past, using the same dataset and the same built-in genome.

I've read in other posts that this could be a memory issue, so I tried setting up a smaller genome (less than 130 kb), and I get the same error.

I can get both of the tools to work properly using files from the history, but this would be much less convenient in the long term.

I would appreciate any ideas on what I could try next.

Thanks,

Samuel

bwa bowtie • 2.0k views
ADD COMMENTlink modified 3.8 years ago • written 3.8 years ago by samuel.tremblay-belzile60
3
gravatar for Daniel Blankenberg
3.8 years ago by
Daniel Blankenberg ♦♦ 1.7k
United States
Daniel Blankenberg ♦♦ 1.7k wrote:

Hi Samuel,

When you rsync'd everything, did you place it all under a /galaxy/data/ directory, or in another directory. The .loc file for bwa, has the path for this set of indexes set at '/galaxy/data/Arabidopsis_thaliana_TAIR10/bwa_index/Arabidopsis_thaliana_TAIR10.fa.bwt', which doesn't seem to appear on your file system. You'll either need to move the data you downloaded there, or update the .loc files to have the path where you saved the downloaded reference data.

ADD COMMENTlink written 3.8 years ago by Daniel Blankenberg ♦♦ 1.7k
1
gravatar for samuel.tremblay-belzile
3.8 years ago by
Canada

I had set up the folders as they were in the .loc files, but since you brought up the path issue, I decided to try adding the complete path in the .loc files instead of starting from my Galaxy folder. This appears to have resolved the issue for BWA, since the task is now running.

Do you have any idea why the Bowtie tool might not be able to detect any of the reference genomes?

Thanks for your help.

ADD COMMENTlink written 3.8 years ago by samuel.tremblay-belzile60

Hello,

Dan may add more but I am curious if you also added the reference genomes to your builds list. This is a required dependency.

Complete help, including data manager and manual installation/index creation plus configuration is here:
http://wiki.galaxyproject.org/Admin/DataIntegration

This wiki has links out to other wiki help specific for builds and data managers, please see the top of the page for these.

Thanks, Jen, Galaxy team

ADD REPLYlink written 3.8 years ago by Jennifer Hillman Jackson25k

I did have the builds list up to date, but I realized that the default tool_data_table_conf.xml file has an entry for BWA, but not for Bowtie. I've added it, and now both tools are working.

Thanks for pointing me in the right direction.

ADD REPLYlink written 3.8 years ago by samuel.tremblay-belzile60

Great, glad this worked. Since Dan's reply helped to get this all going, it would be super if you could mark it as accepted (click on the checkbox next to the reply). This helps others to locate accurate Q/A, and other feedback will be seen by the readers, should they also have follow-up issues to resolve. Take care, Jen, Galaxy team

ADD REPLYlink written 3.8 years ago by Jennifer Hillman Jackson25k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 139 users visited in the last hour