Fetching Ids With Unknown Species From Ensembl Using Python

Question

[jergosh]

Is there a PyCogent (or more generally, Python) equivalent of get_species_and_object_type() from the Ensembl Perl API? I would like to fetch sequences for Ensembl IDs where I don't know what species a given ID comes from. As far as I can see, querying with cogent.db.ensembl requires creating a Genome object for a specific organism, and the Compara instances are also accessed as compara.SpeciesName. Is there a species-agnostic way of performing queries?

I tried using the REST API but it only supports the current release of ensembl whereas the IDs I'm dealing with are from a past release.

Answer 1

[Biojl]

You may want to take a look to REST http://beta.rest.ensembl.org/

You will find it much easier than PyCogent and works with python too.

[jergosh]

The REST API is great but only supports the current version of Ensembl -- I should've said I'm working with IDs from 2011 (now edited my question to reflect that).

Answer 2

[Emily]

This Perl script does exactly what you need it to:

use strict;
use warnings;

use Bio::EnsEMBL::Registry;
my $reg = "Bio::EnsEMBL::Registry";

$reg->load_registry_from_db(
   -host => 'ensembldb.ensembl.org',
   -user => 'anonymous',
);

my $stable_id = 'ENST00000326632';

my ( $species, $object_type, $db_type ) = $reg->get_species_and_object_type($stable_id);

my $adaptor = $reg->get_adaptor( $species, $db_type, $object_type );

my $object = $adaptor->fetch_by_stable_id($stable_id);

print ">", $stable_id, "\n", $object->seq->seq, "\n\n";

I'm really not sure why you're trying to make things more complicated for yourself by avoiding Perl.

[jergosh]

It would maybe make sense to use Perl if this was the only kind of analysis I was doing. As it is, the rest of my code is in Python and my environment is not set up for Perl (no BioPerl, no Ensembl API etc.)