[galaxy-user] Intersect BED data

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

[galaxy-user] Intersect BED data

Stephen Taylor
Hi,

I recently did an intersect on two *identical* BED format feature sets exported from UCSC in Galaxy and got no results (see attached png). I tried it on our local install and remote install and both gave the same answer.

Please can someone advise why this doesn't work?

Many thanks for any help,

Steve



_______________________________________________
galaxy-user mailing list
[hidden email]
http://lists.bx.psu.edu/listinfo/galaxy-user

IntersectGalaxy.PNG (50K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: [galaxy-user] Intersect BED data

Guruprasad Ananda
Hi Steve,

I'm afraid I'm unable to reproduce this scenario with my datasets. Can you please share your history with me ([hidden email]) so that I can take a look at your input datasets?

Thanks,
Guru
Galaxy team.

On Jun 1, 2010, at 9:58 AM, Steve Taylor wrote:

> Hi,
>
> I recently did an intersect on two *identical* BED format feature sets exported from UCSC in Galaxy and got no results (see attached png). I tried it on our local install and remote install and both gave the same answer.
>
> Please can someone advise why this doesn't work?
>
> Many thanks for any help,
>
> Steve
>
>
> <IntersectGalaxy.PNG>_______________________________________________
> galaxy-user mailing list
> [hidden email]
> http://lists.bx.psu.edu/listinfo/galaxy-user


_______________________________________________
galaxy-user mailing list
[hidden email]
http://lists.bx.psu.edu/listinfo/galaxy-user
Reply | Threaded
Open this post in threaded view
|

Re: [galaxy-user] Intersect BED data

Stephen Taylor
Hi Guru,

> I'm afraid I'm unable to reproduce this scenario with my datasets. Can you please share your history with me ([hidden email]) so that I can take a look at your input datasets?
>    

  I have reran a similar analysis again (this time, intersecting a data
set with itself) on the public server and shared it with you.

Thanks!

Steve

> Thanks,
> Guru
> Galaxy team.
>
> On Jun 1, 2010, at 9:58 AM, Steve Taylor wrote:
>
>    
>> Hi,
>>
>> I recently did an intersect on two *identical* BED format feature sets exported from UCSC in Galaxy and got no results (see attached png). I tried it on our local install and remote install and both gave the same answer.
>>
>> Please can someone advise why this doesn't work?
>>
>> Many thanks for any help,
>>
>> Steve
>>
>>
>> <IntersectGalaxy.PNG>_______________________________________________
>> galaxy-user mailing list
>> [hidden email]
>> http://lists.bx.psu.edu/listinfo/galaxy-user
>>      
>    

_______________________________________________
galaxy-user mailing list
[hidden email]
http://lists.bx.psu.edu/listinfo/galaxy-user
Reply | Threaded
Open this post in threaded view
|

Re: [galaxy-user] Intersect BED data

Guruprasad Ananda
Hi Steve,

The problem was with your strand column. Col 6 was assigned as the strand column, but it didn't contain valid strands (+/-). The intersect tool was therefore skipping all the input intervals as invalid. The tool should have displayed a message about invalid lines - we'll get this fixed soon.
In the meantime, you can get the tool to run as follows:
Please click on the pencil icon next to your input dataset and uncheck the checkbox next to strand column and save this change. Rerunning intersect tool should now work just fine. The same is shown in this history: http://main.g2.bx.psu.edu/u/guru/h/clone-of-unnamed-history-shared-by-stephentaylorimmoxacuk-active-items-only

Hope this answers your question.
Thanks for using Galaxy,
Guru.



On Jun 1, 2010, at 3:44 PM, Stephen Taylor wrote:

> Hi Guru,
>
>> I'm afraid I'm unable to reproduce this scenario with my datasets. Can you please share your history with me ([hidden email]) so that I can take a look at your input datasets?
>>  
>
> I have reran a similar analysis again (this time, intersecting a data set with itself) on the public server and shared it with you.
>
> Thanks!
>
> Steve
>> Thanks,
>> Guru
>> Galaxy team.
>>
>> On Jun 1, 2010, at 9:58 AM, Steve Taylor wrote:
>>
>>  
>>> Hi,
>>>
>>> I recently did an intersect on two *identical* BED format feature sets exported from UCSC in Galaxy and got no results (see attached png). I tried it on our local install and remote install and both gave the same answer.
>>>
>>> Please can someone advise why this doesn't work?
>>>
>>> Many thanks for any help,
>>>
>>> Steve
>>>
>>>
>>> <IntersectGalaxy.PNG>_______________________________________________
>>> galaxy-user mailing list
>>> [hidden email]
>>> http://lists.bx.psu.edu/listinfo/galaxy-user
>>>    
>>  
>


_______________________________________________
galaxy-user mailing list
[hidden email]
http://lists.bx.psu.edu/listinfo/galaxy-user
Reply | Threaded
Open this post in threaded view
|

Re: Intersect BED data

Stephen Taylor
Hi Guru,

>
> The problem was with your strand column. Col 6 was assigned as the strand column, but it didn't contain valid strands (+/-). The intersect tool was therefore skipping all the input intervals as invalid. The tool should have displayed a message about invalid lines - we'll get this fixed soon.
> In the meantime, you can get the tool to run as follows:
> Please click on the pencil icon next to your input dataset and uncheck the checkbox next to strand column and save this change. Rerunning intersect tool should now work just fine. The same is shown in this history: http://main.g2.bx.psu.edu/u/guru/h/clone-of-unnamed-history-shared-by-stephentaylorimmoxacuk-active-items-only
>

I see. That's a bit of a gotcha, especially for novice users/unknown data sets. Maybe the strand and name columns should be unchecked as default?

Thanks for your help.

Steve

> Hope this answers your question.
> Thanks for using Galaxy,
> Guru.
>
>
>
> On Jun 1, 2010, at 3:44 PM, Stephen Taylor wrote:
>
>> Hi Guru,
>>
>>> I'm afraid I'm unable to reproduce this scenario with my datasets. Can you please share your history with me ([hidden email]) so that I can take a look at your input datasets?
>>>
>>
>> I have reran a similar analysis again (this time, intersecting a data set with itself) on the public server and shared it with you.
>>
>> Thanks!
>>
>> Steve
>>> Thanks,
>>> Guru
>>> Galaxy team.
>>>
>>> On Jun 1, 2010, at 9:58 AM, Steve Taylor wrote:
>>>
>>>
>>>> Hi,
>>>>
>>>> I recently did an intersect on two *identical* BED format feature sets exported from UCSC in Galaxy and got no results (see attached png). I tried it on our local install and remote install and both gave the same answer.
>>>>
>>>> Please can someone advise why this doesn't work?
>>>>
>>>> Many thanks for any help,
>>>>
>>>> Steve
>>>>
>>>>
>>>> <IntersectGalaxy.PNG>_______________________________________________
>>>> galaxy-user mailing list
>>>> [hidden email]
>>>> http://lists.bx.psu.edu/listinfo/galaxy-user
>>>>
>>>
>>
>

_______________________________________________
galaxy-user mailing list
[hidden email]
http://lists.bx.psu.edu/listinfo/galaxy-user