[ensembl-dev] "subseq" error

[Will McLaren]

Hello again,

I've patched VariationFeatureAdaptor.pm so it now works with 'ins'
types and should return a sensible error message if it fails. You can
update the file from CVS or re-download the tar file if you aren't
using CVS.

It will still not work with NC_000021.8 (now it will warn you instead
of crashing), but if you replace this with 21 then it should work.

Thanks

Will

On 29 February 2012 10:10, S Venkata Suresh Kumar
<suresh.surampudi at india.semanticbits.com> wrote:
> Thanks. I will be looking forward till all the consequences are
> incorporated in HGVS syntax.
>
> However, VEP it self is very useful tool so far for us.
>
> regards
> vs
> On 02/29/2012 03:10 PM, Will McLaren wrote:
>> Hello again,
>>
>> Currently the HGVS parser only works reliably with a limited subset of
>> the HGVS nomenclature types. The 'ins' type is not currently
>> supported.
>>
>> Furthermore, when using --refseq only coding HGVS notations can be
>> parsed - we don't store the chromosomes as you have supplied here
>> (e.g. NC_000021.8:g.36421412_36421413insA).
>>
>> You could, however, easily just use the chromosome name and not use
>> --refseq here e.g.
>>
>> 21:g.36421412_36421413insA
>>
>> however, this unfortunately will still not work as we cannot yet parse
>> the 'ins' type. Other basic types should work correctly, however, and
>> we will work on getting the other types supported in future.
>>
>> However, I am still surprised that the script does not fail gracefully
>> - when I try this I get the error:
>>
>> WARNING: Unable to parse HGVS notation 'NC_000021.8:g.36421412_36421413insA'
>> The variant class for HGVS notation
>> 'NC_000021.8:g.36421412_36421413insA' is unknown or could not be
>> correctly recognized at
>> ensembl-variation/modules/Bio/EnsEMBL/Variation/DBSQL/VariationFeatureAdaptor.pm
>>
>> Thanks
>>
>> Will
>>
>> On 29 February 2012 08:06, S Venkata Suresh Kumar
>> <suresh.surampudi at india.semanticbits.com> wrote:
>>> I am using a file (without header) with one HGVS syntax :
>>> NC_000021.8:g.36421412_36421413insA
>>>
>>> I am trying to convert this into VCF (to test vep script). Following is
>>> the way I am trying:
>>>
>>> $ perl /opt/vep/variant_effect_predictor.pl -i hgvs -o hgvs.vcf -refseq
>>> -convert vcf --verbose -force
>>>
>>> Regards
>>> vs
>>> On 02/28/2012 10:55 PM, Will McLaren wrote:
>>>> Hello,
>>>>
>>>> Are you able to provide an example of the line that is causing this issue?
>>>>
>>>> Thanks
>>>>
>>>> Will McLaren
>>>> Ensembl Variation
>>>>
>>>> On 28 February 2012 11:13, S Venkata Suresh Kumar
>>>> <suresh.surampudi at india.semanticbits.com> wrote:
>>>>> Hi All
>>>>>
>>>>> When i am trying to convert a hgvs file to vcf file, I am getting
>>>>> following error:
>>>>>
>>>>> "Can't call method "subseq" on an undefined value at
>>>>> /opt/vep/ensembl-variation/modules/Bio/EnsEMBL/Variation/DBSQL/VariationFeatureAdaptor.pm
>>>>> line 1272, <GEN0> line 1."
>>>>>
>>>>> any clues ?
>>>>>
>>>>> I am using vep 2.4, OS linux, bioperl 1.6.901-2.
>>>>>
>>>>> regards
>>>>> vs
>>>>>
>>>>> _______________________________________________
>>>>> Dev mailing list    Dev at ensembl.org
>>>>> List admin (including subscribe/unsubscribe): http://lists.ensembl.org/mailman/listinfo/dev
>>>>> Ensembl Blog: http://www.ensembl.info/
>>>> _______________________________________________
>>>> Dev mailing list    Dev at ensembl.org
>>>> List admin (including subscribe/unsubscribe): http://lists.ensembl.org/mailman/listinfo/dev
>>>> Ensembl Blog: http://www.ensembl.info/
>>> _______________________________________________
>>> Dev mailing list    Dev at ensembl.org
>>> List admin (including subscribe/unsubscribe): http://lists.ensembl.org/mailman/listinfo/dev
>>> Ensembl Blog: http://www.ensembl.info/
>> _______________________________________________
>> Dev mailing list    Dev at ensembl.org
>> List admin (including subscribe/unsubscribe): http://lists.ensembl.org/mailman/listinfo/dev
>> Ensembl Blog: http://www.ensembl.info/
>
> _______________________________________________
> Dev mailing list    Dev at ensembl.org
> List admin (including subscribe/unsubscribe): http://lists.ensembl.org/mailman/listinfo/dev
> Ensembl Blog: http://www.ensembl.info/