[ensembl-dev] VEP Rest API very slow

[Sarah Hunt]

Hi Beat,

Don't worry about changing your code. We have established that the 
problem is due to a lack of indexing on some of the reference data VEP 
is using. We will fix this as soon as possible.

If you need to use the service right now, the reference files containing 
both Ensembl and RefSeq transcripts are indexed, so it would be quicker 
to use a request like this:

https://rest.ensembl.org/vep/hsapiens/id/rs539044217?merged=1&content-type=application/json

Apologies for the inconvenience.

Best wishes,

Sarah


On 19/07/2018 17:15, Wolf Beat wrote:
> Dear Sarah,
>
>
> Thank you for the fast answer. As there are incompatible changes between e93 and e92, notably in the VEP endpoint, moving back to e93 now that the my code has changed, is not ideal.
>
> What is the plan for the VEP backend, will it be broken for the forseable future (which means i need to revert my code changes and do some other changes), or is it a matter of days, in which case i can avoid changing my code?
>
>
> Kind regards
>
>
> Beat Wolf
>
> ________________________________
> From: Dev <dev-bounces at ensembl.org> on behalf of Sarah Hunt <seh at ebi.ac.uk>
> Sent: Thursday, July 19, 2018 5:41:13 PM
> To: Ensembl developers list
> Subject: Re: [ensembl-dev] VEP Rest API very slow
>
>
> Hi Beat,
>
> We are looking into this, but in the meantime, the archive of our e!92
> release is available here:  http://wp-p1m-6e.ebi.ac.uk:5000
>
> It should be available as: http://e92.rest.ensembl.org/ shortly.
>
> Sarah
>
> On 19/07/2018 16:08, Wolf Beat wrote:
>> Hello,
>>
>>
>> i was wondering why the ensembl VEP rest API is very slow.
>>
>> When i try for example this query:
>>
>> http://rest.ensembl.org/vep/hsapiens/id/rs539044217?content-type=application/json<http://rest.ensembl.org/vep/hsapiens/id/rs539044217?distance=-100000&content-type=application/json>
>>
>>
>> It takes up to 10 seconds.
>>
>> It seems like the issue is not related to the actual VEP work (effect prediction), because if i add the following parameter:
>>
>> http://rest.ensembl.org/vep/hsapiens/id/rs539044217?distance=-100000&content-type=application/json
>>
>>
>> thus removing basically all genes from the prediction, it takes an equal amount of time.
>>
>>
>> This is quite an urgent issue for me. Is there a faster way to retrieve frequences (ok that one i know) AND polyphen/sift predictions in the same query, that is faster?
>>
>>
>> Kind regards
>>
>>
>> Beat Wolf
>>
>>
>> _______________________________________________
>> 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/
> _______________________________________________
> 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/
> _______________________________________________
> 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/