That's a neat trick, and I definitely wouldn't have thought of that approach, so thanks for that!

After I finished writing this out, I realized it was super long. So here are the questions I'm asking up front, so you can choose whether or not to read the details. Thanks!

1. How do I output the quality scores when converting from FASTQ to FASTA?

2. Does the SAM-to-interval tool output only mapped reads by looking at the flag values?

3. Why am I getting the mentioned error and is there a way to resolve it?

Here are the details:

I don't see an option to output both the sequences and the quality scores. I found two FASTQ-to-FASTA converters (one under the "Convert Formats" and the other in the FastX Toolkit) and both only output one fasta file with the sequences. Am I missing something, or should I be using some other tool to output both the sequences and the quality scores?
The Extract Genomic Sequences tool seems to want an Interval file as input, not a list of IDs. Does that mean I should convert the filtered SAM output to Interval? Currently I'm using the SAM-to-interval conversion to extract the mapped reads and make the data more manageable in one step (pretty sure I picked that up from one of the tutorials...). I was assuming that by definition it could only output an interval if it was mapped, and if so, I wouldn't be able to convert the unmapped reads to Interval anyway. Is that wrong?
I was setting up a workflow with Bowtie and I noticed that the Workflow Editor does show options to output unmapped reads. But when I try to output them, I get this error:

"Error due to input mapping of 'Compute quality statistics' in 'output_unmapped_reads_l'. A common cause of this is conditional outputs that cannot be determined until runtime, please review your workflow."

Superficially, this seems silly. Obviously a "conditional output" will not be determined until runtime because it's dependent on something else. So why is that an error? I have tried outputting to a few different tools, so it doesn't seem to be specific to the tool into which the unmapped reads go (in this case, "Compute Quality Statistics").

Any thoughts, insights, or even other approaches to the original problem would be great. Currently, I'm thinking my best bet is to filter out the unmapped reads locally with a Perl script and re-upload, but that felt like overkill and time-consuming when I will inevitably want to tweak or re-run things. Also, installing a local instance is currently not an option for me (though it should be in a few months). In any case, I appreciate your help a lot!

Thanks, again!

Mayank Tandon

On Thu, Jul 18, 2013 at 5:38 PM, Jennifer Jackson <jen@bx.psu.edu> wrote:

Hi Mayank,

The best option I know of is to do the following:

1 - obtain the sequence identifiers for the unmapped reads by filter the SAM file, then cutting them out
2 - convert the original FASTQ file to FASTA - you should get two output, one for the sequences and one for the quality score values
3 - use the tool "Fetch Sequences -> Extract Genomic DNA". The query is the list from #1, the target is the "genome" from #2. Do this twice - once for seqs, once for quals. This means using the target datasets from #2 as Custom Reference genomes - help about how to do this is here:
http://wiki.galaxyproject.org/Support#Custom_reference_genome
4 - combine the FASTA seq and qual files back to FASTQ

If you will be doing this again, then capture the process into a workflow for future use, in a way creating your own "tool".

Hopefully this helps!

Jen
Galaxy

On 7/18/13 9:40 AM, Mayank Tandon wrote:
I was hoping this question had been asked, but I haven't been able to find it. I want to output the unmapped reads from bowtie as a fastq file for subsequent mapping to other genomes (i.e. the "--un <filename>" option). I know I can extract the unmapped reads by filtering on the bitwise values in the sam output and converting to fastq with the Picard tool, but I'm using colorspace data and bowtie converts them to letterspace. My understanding (coming mostly from forums and personal discussions) was that the color-to-letter conversion was somehow lossy so mapping the colorspace data directly is always preferable.

So the question is: Is bowtie's '--un' option implemented in Galaxy and if so, how do I access it?

Thanks in advance!

Mayank Tandon
___________________________________________________________
The Galaxy User list should be used for the discussion of
Galaxy analysis and other features on the public server
at usegalaxy.org.  Please keep all replies on the list by
using "reply all" in your mail client.  For discussion of
local Galaxy instances and the Galaxy source code, please
use the Galaxy Development list:

  http://lists.bx.psu.edu/listinfo/galaxy-dev

To manage your subscriptions to this and other Galaxy lists,
please use the interface at:

  http://lists.bx.psu.edu/

To search Galaxy mailing lists use the unified search at:

  http://galaxyproject.org/search/mailinglists/
-- Jennifer Hillman-Jackson Galaxy Support and Training http://galaxyproject.org