Hi Dan,

The problem is that your FASTA file has a blank lines in it (the first is line 448). You can get rid of them within Galaxy with the Select tool (under Filter and Sort). Select the NOT Matching option and enter "^$" (without the quotes) into the pattern box. This will remove all lines that have nothing between the start (^) and end ($) of the line.

Note that I tested SAM-to-BAM both with and without re-mapping with the corrected FASTA file, and the two resulting BAM files were slightly different in size, so you may need to remap.


On Jan 10, 2011, at 4:02 PM, da...@pitt.edu wrote:


We're loving Galaxy.  This is an excellent interface and set of tools.

The issue we're having is running SAM to BAM. We've been following the
"Illumina Mapping: Single Reads" screencast, with the only significant
difference (I think) being that we're using our own uploaded reference
genomes instead of a pre-built index for the Bowtie alignment.

When we get to the SAM to BAM step, we have two options for "Choose the
source for the reference list": 1. Locally Cached, 2. History.

1. If we try to use "Locally Cached", we get the following error before the job is even added to the queue: "Unspecified genome build, click the pencil icon in the history item to set the genome build". We've tried to follow those directions, but don't see the reference genome we want in the

2. If we try to use "History" and specify the SAM file and then the FASTA
file we used to do the bowtie alignment, we get the following error:

An error occurred running this job: Samtools Version: 0.1.12 (r862) Error
creating indexes from reference
(/galaxy/home/g2main/galaxy_main/database/files/001/894/ dataset_1894598.dat),
[fai_build_core] different line length in sequence
Segmentation fault

Any guidance would be wonderful.

My user name is: da...@pitt.edu

Thanks for your help,
Dan Russell
University of Pittsburgh

galaxy-user mailing list

galaxy-user mailing list

Reply via email to