interval format from pileup

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

interval format from pileup

Katie Hyma
Hi all,

I ran into this problem, and thought others might run in to it as well:
When using bowtie and samtools, positions are formatted with a 1-base offset, whereas interval format requires 0-base offset.
Recently, I used the 'Filter pileup on coverage and SNPs' tool with the 'convert coordinates to intervals' option, and used the output with the 'aaChanges' tool to detect amino acid changes.
In the aaChanges output there was an off-by-one discrepancy with a previous analysis, which brought this problem to my attention.
Should the 'Filter pileup' tool be changed to produce output with 0-base offset when interval format is selected, since interval format is defined as having 0-base output? It seems that others might run into the same problem, and it would be an easy one to miss.
Although the pileup format is now deprecated, the same problem might occur if/when the change to VCF format occurs, as it also uses 1-base offset. It also may be a problem for other tools that use interval format as an input, expecting a 0-based position.

Best,
Katie

___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/
Reply | Threaded
Open this post in threaded view
|

Re: interval format from pileup

Anton Nekrutenko
Katie:
"Filter pileup" outputs 0-based coordinates if you use the "Convert coordinates to intervals" option. So the issue may be related to aaChanges tools. Can you give us a specific example, so we can track things down.

Thanks!

anton
galaxy team



Anton Nekrutenko
http://nekrut.bx.psu.edu
http://usegalaxy.org



On May 12, 2011, at 10:37 AM, Katie Hyma wrote:

> Hi all,
>
> I ran into this problem, and thought others might run in to it as well:
> When using bowtie and samtools, positions are formatted with a 1-base offset, whereas interval format requires 0-base offset.
> Recently, I used the 'Filter pileup on coverage and SNPs' tool with the 'convert coordinates to intervals' option, and used the output with the 'aaChanges' tool to detect amino acid changes.
> In the aaChanges output there was an off-by-one discrepancy with a previous analysis, which brought this problem to my attention.
> Should the 'Filter pileup' tool be changed to produce output with 0-base offset when interval format is selected, since interval format is defined as having 0-base output? It seems that others might run into the same problem, and it would be an easy one to miss.
> Although the pileup format is now deprecated, the same problem might occur if/when the change to VCF format occurs, as it also uses 1-base offset. It also may be a problem for other tools that use interval format as an input, expecting a 0-based position.
>
> Best,
> Katie
> ___________________________________________________________
> Please keep all replies on the list by using "reply all"
> in your mail client.  To manage your subscriptions to this
> and other Galaxy lists, please use the interface at:
>
>  http://lists.bx.psu.edu/


___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/
Reply | Threaded
Open this post in threaded view
|

Re: interval format from pileup

Anton Nekrutenko
Thanks for letting me know. I'm also forwarding this galaxy-dev, so the thread in complete.

Thanks,

anton



On May 12, 2011, at 2:30 PM, Katie Hyma wrote:

Hi Anton,

You're right, sorry for the false alarm. After a more careful inspection of my data I'm pleased to say that I was wrong and that everything with galaxy is fine, it turns out that the company that generated the previous data had generated the error; good think I checked those results, I had just assumed I that my analysis had the error!

Katie

On Thu, May 12, 2011 at 10:54 AM, Anton Nekrutenko <[hidden email]> wrote:
Katie:
"Filter pileup" outputs 0-based coordinates if you use the " Convert coordinates to intervals" option. So the issue may be related to aaChanges tools. Can you give us a specific example, so we can track things down.

Thanks!

anton
galaxy team



Anton Nekrutenko
http://nekrut.bx.psu.edu
http://usegalaxy.org



On May 12, 2011, at 10:37 AM, Katie Hyma wrote:

> Hi all,
>
> I ran into this problem, and thought others might run in to it as well:
> When using bowtie and samtools, positions are formatted with a 1-base offset, whereas interval format requires 0-base offset.
> Recently, I used the 'Filter pileup on coverage and SNPs' tool with the 'convert coordinates to intervals' option, and used the output with the 'aaChanges' tool to detect amino acid changes.
> In the aaChanges output there was an off-by-one discrepancy with a previous analysis, which brought this problem to my attention.
> Should the 'Filter pileup' tool be changed to produce output with 0-base offset when interval format is selected, since interval format is defined as having 0-base output? It seems that others might run into the same problem, and it would be an easy one to miss.
> Although the pileup format is now deprecated, the same problem might occur if/when the change to VCF format occurs, as it also uses 1-base offset. It also may be a problem for other tools that use interval format as an input, expecting a 0-based position.
>
> Best,
> Katie
> ___________________________________________________________
> Please keep all replies on the list by using "reply all"
> in your mail client.  To manage your subscriptions to this
> and other Galaxy lists, please use the interface at:
>
>  http://lists.bx.psu.edu/




___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/