[ensembl-dev] REST server 3.0 vep endopoint works only with human specie

mag mr6 at ebi.ac.uk
Wed Jul 23 10:41:25 BST 2014


Hi Paolo,

The variation provides a vep cache for all species with variation data
ftp://ftp.ensembl.org/pub/release-75/variation/VEP/

This cache contains all known variations, ordered by location.
Combined with a fasta file for the genomic sequence, this means we can 
access all the required information from file rather than querying the 
database every time.

With single queries, the performance is comparable.
If you are running multiple queries, especially if you are working on 
the same region, the file retrieval is faster as it can re-use indexed data.


Regards,
Magali

On 23/07/2014 10:14, Paolo Cozzi wrote:
>
> Dear Magali,
>
> Thank you for your replay and your support. Could you describe briefly 
> vep cache? is something like memcached?
>
> At the moment, I don't think that I need faster performance on other 
> species, but I will remember your suggestion.
>
> Regards,
>
> Paolo.
>
>
>
> Il 22/07/2014 18:36, mag ha scritto:
>> Hi Paolo,
>>
>> Thank you for reporting this and apologies for the error.
>>
>> We were wrongly using some human-specific configuration for all 
>> species, this has now been fixed on the live server.
>>
>> If you are using POST requests for VEP, please be aware that only 
>> human uses the vep cache.
>> For other species, we are still using direct database connection, 
>> which will result in lower performance.
>>
>> If you are interested in getting faster POST support for other 
>> species, please contact us.
>>
>>
>> Regards,
>> Magali
>>
>> On 22/07/2014 11:55, Paolo Cozzi wrote:
>>>
>>> Dear developer,
>>>
>>> I found an odd response from the vep/:species/region/ 
>>> <http://rest.ensembl.org/documentation/info/vep_region_post> of the 
>>> new ensembl REST server ( http://rest.ensembl.org/) while searching 
>>> for species other than human. For example, if I do 2 POST request 
>>> with curl on human and cow (the sample data are the same as in the 
>>> VEP webserver):
>>>
>>> $ curl -H "content-type:application/json" -H 
>>> "accept:application/json" --data '{ "variants" : ["1  909238 909238  
>>> G/C  +", "3  361464  361464  A/-  +", "5  121187650 121188519  DUP"] 
>>> }' http://rest.ensembl.org/vep/human/region > output_human.txt
>>>
>>> $ curl -H "content-type:application/json" -H 
>>> "accept:application/json" --data '{ "variants" : ["1  909238 909238  
>>> G/C  +", "3  361464  361464  A/-  +", "5  121187650 121188519  DUP"] 
>>> }' http://rest.ensembl.org/vep/cow/region > output_cow.txt
>>>
>>> I will find the same results (diff give me no output), even VEP 
>>> gives me 2 different results. Could be a problem of variation endpoint?
>>>
>>> Thanks for your support,
>>>
>>> best regards,
>>>
>>> Paolo.
>>>
>>>
>>>
>>> _______________________________________________
>>> 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 listDev 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/20140723/f80b2a77/attachment.html>


More information about the Dev mailing list