Stitch MAF blocks - can gaps in reference be kept?

Question: Stitch MAF blocks - can gaps in reference be kept?

3.8 years ago by

France

kjsiddle • 30 wrote:

Dear all,

I am using the galaxy tool "Stitch MAF blocks" to extract multiple alignments (using the 28-way multiZ) and concatenate them into a single sequence per species. I have noticed that the resulting hg18 sequence in the fasta file contains no gaps (even though there are gaps within the individual blocks). For my downstream applications I need to keep these gaps.

Does anyone know if there is a way with this tool to keep the gaps? Alternatively, can anyone suggest another approach that might solve this problem.

Many thanks in advance,

Katherine

galaxy • 1.3k views

ADD COMMENT • link •

modified 3.8 years ago by Jennifer Hillman Jackson ♦ 25k • written 3.8 years ago by kjsiddle • 30

3.8 years ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Hello,

Please review the tool option "Split into Gapless MAF blocks:" near the bottom of the form. Currently, this is the method to preserve any gaps that may occur with this tool in the primary genome.

Best, Jen, Galaxy team

ADD COMMENT • link written 3.8 years ago by Jennifer Hillman Jackson ♦ 25k

Thank you for your response. I have actually already tried this option and there was no difference (i.e. no gaps in hg18) when I used both "yes" and "no". Just as an example, one of the regions I am trying to extract is this one: chr12(+):9111571-9111685.

When I use "stitch" I get the following:

>hg18.chr12(+):9111571-9111685
ACATTGACCAGAAAAAGTGTTTATTCATCAAGTCTTTAAAGATACAAAAACACGTGTCTTCTGTGGAGCTCTGAGAACAGGACTCCAGCAAAGCACTTTTCAGCCTTGTGGTCT

However, if I use extract MAF blocks followed by MAF to fasta I get:

>hg18
ACATT---------GACC-AGAAAAAGTGTTTATTCATCAAGTCTTT----------------------AAAGATAC--AAAAACA--CGTGTCTTCTGTGGAGCTCTGAGAACAGGACT-CCAGCAAAGCACTTTTCAGCCTTGTGGTCT

I could use the latter option, however, it has two principle limitations. 1) For sequences on the negative strand additional processing is needed, which seems to be automatically handled by stitch, and 2) I would need to do extract and convert each sequence individually as MAF to Fasta will concatenate all sequences in the file. As I am searching for around 10,000 sequences genome-wide this doesn't seem tractable, unless there is a way to loop over a list of queries in Galaxy.

The tools on galaxy are great for handling what is a pretty difficult file type, but I am just stuck on this one, simple, thing! Any suggestions for how to get around it will be greatly appreciated.

Many thanks,

Katherine

ADD REPLY • link modified 3.8 years ago • written 3.8 years ago by kjsiddle • 30

Hello,

This is how the tools function. But you can use a combination of other tools to extract the sequences directly from the MAF output and reverse compliment any reported on the (-) strand. I shared an example workflow (below) that will do this in batch for the entire output from the "Extract MAF Blocks" tool. I'll leave it shared for a few days until you let me know that you have it.

Once you have a copy, you can edit it any way you want. For example, if you wanted to save more information from the MAF line, you can add in a step like "Text Manipulation -> Merge Columns" once the data is converted to a tabular format, before running the other steps. Or get very fancy and "Text Manipulation -> Add column" with an underscore, and use that underscore between the fields when running Merge, to make the sequence identifier more readable. This will not be the same sequence identifier as "Stitch MAF Blocks" produces, but may be enough.

https://usegalaxy.org/u/jen-bx-galaxy-edu/w/maf-sequences

Thanks! Jen, Galaxy team

ADD REPLY • link written 3.8 years ago by Jennifer Hillman Jackson ♦ 25k

Hello,

I am interested in a very similar problem to the original poster. But I can't seem to access the workflow that you have posted?

Would you be able to share it with me please?

Many thanks,

Thomas

ADD REPLY • link written 2.9 years ago by Thomas • 0

Hi Thomas, Let me see if I still have this. If not, will create a permanent version then share it back here. Jen, Galaxy team

ADD REPLY • link written 2.9 years ago by Jennifer Hillman Jackson ♦ 25k

Hello again. I created an example workflow under Shared Data: Published Workflows. This can be imported and edited to suit different analysis needs by everyone.

https://usegalaxy.org/u/jen/w/bed-to-maf-to-fasta

Hopefully this helps! Jen, Galaxy team

ADD REPLY • link written 2.9 years ago by Jennifer Hillman Jackson ♦ 25k

Thank you, much appreciated

Thomas

ADD REPLY • link written 2.9 years ago by Thomas • 0

Similar posts • Search »