[ensembl-dev] VEP Flanking sequence plugin

Mon Jun 3 11:27:20 BST 2013

Hello,

My plugin start giving errors.

ERROR: Forked process failed
Plugin 'flanking_sequence' went wrong: *Can't call method "expand" on an 
undefined value at 
/share/apps/scripts/genomics/global/annotation/ensembl_71_genetonic/vep_config/Plugins/flanking_sequence.pm 
line 72.*

I've tried to fix the code adding if(defined ...) but seems to have no 
effect. I didn't see this bug until today.

How can I check if the value of any of the two expand feature_slice 
really exists?
I thought this would exist for all features. So I suppose I'm wrong and 
what is causing the plugin to fail obtaining the flanking sequence is 
the expand method call but I've no idea on how to fix it.

Here's plugin code updated: 
https://github.com/guillermomarco/vep_plugins_71/blob/master/flanking_sequence.pm

Regards,
Guillermo.

On 04/24/2013 12:11 PM, Guillermo Marco Puche wrote:
> Hello,
>
> Thank you for information.
> I was already running VEP with --check_existing flag but i had no idea 
> on how to retrieve the rsID (I was looking the Variation API so i 
> thought that
> $tva->variation_feature->variation_name;
>
> would be enough.
>
> As you said I just wanted reference sequence on either side of the 
> variant so not even rsID needed.. but it's good to know how I can 
> obtain it for a near future :)
>
> Just to prevent from asking these kind of question, how can i print 
> all the content from the $va object for example?
>
> I mean how can I know this /"you will then find a simplified hash 
> representing each overlapping variant in the array @{$vf->{existing}} 
> (where $vf = $tva->variation_feature)."/ I can't see a "existing" 
> method in the API, I suppose that is comming from the own VEP script./
>
> /
> I also changed the (I don't know if I should undo this change)
> sub feature_types {
>      return ['Transcript', 'Feature'];
> }
> I was playing around this but the information on VEP script web 
> doesn't clarify this too much. It has to do with handlers and 
> overwriting but I don't understand very well.
>
> Updated script in repo if anyone wants to use it.
>
> Thank you very much Will.
>
> Best regards,
> Guillermo./
>
> /I don't want to spam list for questions like that, but I can't find that
> On 04/24/13 10:50, Will McLaren wrote:
>> Hello,
>>
>> A variation feature object is created by the VEP to represent each
>> line of your input. If you don't give it a name in your input, then
>> the VEP constructs a name from the coordinates, which is what you are
>> seeing. Names can be added for variants in the third column (VCF) or
>> sixth column (tab-delimited input).
>>
>> If you want the rsID of any existing variants that overlap yours, you
>> must run the VEP with the --check_existing flag; you will then find a
>> simplified hash representing each overlapping variant in the array
>> @{$vf->{existing}} (where $vf = $tva->variation_feature).
>>
>> If you are looking for the original flanking sequence submitted with
>> the existing variant to dbSNP, then we don't store this in the
>> database. However, if you just want to retrieve the reference sequence
>> either side of your variant, that is easily done by getting and
>> expanding the feature slice attached to the variant (here I'm getting
>> 100bp either side):
>>
>> my $five_prime_seq = $vf->feature_Slice->expand(100, -1)->seq;
>> my $three_prime_seq = $vf->feature_Slice->expand(-1, 100)->seq;
>>
>> Regards
>>
>> Will
>>
>>
>>
>>
>> On 24 April 2013 09:13, Guillermo Marco Puche
>> <guillermo.marco at sistemasgenomicos.com>  wrote:
>>> Hello,
>>>
>>> I updated the code in my git repo, and I still can't make it
>>> work:https://github.com/guillermomarco/vcf_input/blob/master/flanking_sequence.pm
>>>
>>> I still don't get why this is returning a string of type 1_41304886_C/T and
>>> not the rsID.
>>>
>>> $tva->variation_feature->variation_name;
>>>
>>>
>>> Thank you.
>>>
>>> Best regards,
>>> Guillermo.
>>>
>>>
>>> On 04/23/13 16:45, Guillermo Marco Puche wrote:
>>>
>>> Hello,
>>>
>>> I'm developing a plugin for VEP that calculates 5' & 3' flanking sequence
>>> for each rs.
>>>
>>> I've the plugin 95% completed. I'm just missing how are rs IDs called on
>>> VEP.
>>> Looking into Enseml Variation API I've found this:
>>> http://www.ensembl.org/info/docs/Doxygen/variation-api/classBio_1_1EnsEMBL_1_1Variation_1_1VariationFeature.html
>>>
>>> It seems that I need VariationFeature adaptor set so I can call
>>> "variation_name" method get the rs and then call "five_prime_flanking_seq" &
>>> "three_prime_flanking_seq" methods to get the flanking sequence.
>>>
>>> I don't know why even after changing adaptor I'm still getting
>>> "2_26739423_T/C" if I'm not using TranscriptVariation adaptor but
>>> VariantFeature I should be getting rsxxxxxx ID.
>>>
>>> Plugin code can be found in Github:
>>> https://github.com/guillermomarco/vcf_input/blob/master/flanking_sequence.pm
>>>
>>>
>>>
>>>
>>> Thank you !
>>>
>>> Best regards,
>>> Guillermo.
>>>
>>>
>>> _______________________________________________
>>> Dev mailing listDev at ensembl.org
>>> Posting guidelines and subscribe/unsubscribe info:
>>> http://lists.ensembl.org/mailman/listinfo/dev
>>> Ensembl Blog:http://www.ensembl.info/
>>>
>>>
>>> _______________________________________________
>>> Dev mailing listDev at ensembl.org
>>> Posting guidelines and subscribe/unsubscribe info:
>>> http://lists.ensembl.org/mailman/listinfo/dev
>>> Ensembl Blog:http://www.ensembl.info/
>>>
>
>
> _______________________________________________
> Dev mailing list    Dev at ensembl.org
> Posting guidelines and subscribe/unsubscribe info: http://lists.ensembl.org/mailman/listinfo/dev
> Ensembl Blog: http://www.ensembl.info/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20130603/6a727b92/attachment.html>