[ensembl-dev] get sequence from different build

[Jan-hinnerk Vogel]

Hi,

it's possible; depending on how you get your database adaptor.  
Registry.pm
provides different methods for that. You can also get a DBAdaptor  
explicitly
and loop trough an array of database names to retrieve different  
DBAdaptors.

My choice would be :

my $dba = Bio::EnsEMBL::DBSQL::DBAdaptor->new(
                                               -dbname => $opt{dbname},
                                               -user => $opt{dbuser},
                                               -pass => $opt{dbpass},
                                               -host => $opt{dbhost},
                                               -port => $opt{dbport},
                                              );

... and then stick it in a for loop and loop over the dbnames .......

Cheers,

           Jan-Hinnerk Vogel


On 16 Sep 2010, at 15:00, mailsvl at fastmail.fm wrote:

> Ok, so this means you cannot get the sequences of different builds  
> (for
> the same chromosomal coordinates) within one script? Or is this  
> possible
> by changing the db version temporarily (with eg reset_DBAdaptor...)?
>
> -Stef
>
>
> ----- Original message -----
> From: "Jan-hinnerk Vogel" <jhv at sanger.ac.uk>
> To: mailsvl at fastmail.fm
> Cc: "Ensembl DevList" <dev at ensembl.org>
> Date: Thu, 16 Sep 2010 14:30:46 +0100
> Subject: Re: [ensembl-dev] get sequence from different build
>
> Hi Stef,
>
> ooops - it seems the tutorial is wrong, sorry. We'll correct it asap.
>
> What happend is that the 'strand' argument, which should be the 5th
> argument, slipped somehow.
> The assembly version should be the 6th argument, not the 5th. A good
> way to get warned about
> those things is to  use perl -w or use 'use warnings' in your script -
> you see something like
>
> "Argument "GRCh37" isn't numeric in integer multiplication (*) at / 
> nfs/
> ensembl/jhv//cvs_checkout//ensembl_live/modules/Bio/EnsEMBL/Mapper.pm
> line 314." so something is wrong ...
>
>
> http://www.ensembl.org/info/docs/Pdoc/ensembl/modules/Bio/EnsEMBL/DBSQL/SliceAdaptor.html#POD15
>
>
> You currently can't fetch sequence data from a different assembly
> version from the same database  - only the mapping is stored,
> not the sequence of older assemblies. If you try it, you should get a
> string of NNNNN's back.
>
> To fetch sequence for GRCh37 use the latest human core :
> homo_sapiens_core_59_37d
> to fetch seq for NCBI36 you have to use an older database and connect
> with an older API checkout - ie homo_sapiens_core_***_36*
>
>
> Hth,
>            Jan-Hinnerk Vogel
>
>
>
> ==
> Ensembl Genebuild Team Project Leader
> www.ensembl.org
>
>
>
> On 16 Sep 2010, at 12:54, mailsvl at fastmail.fm wrote:
>
>> Hi,
>>
>> How can I get a slice from a different build then the default? When I
>> try to compare GCRh37 and NCBI36 I keep getting the same sequences
>> (from
>> GCRh37), no matter what region I try.
>
> I assume you mean i "GRCh37" ?
>
>>
>> The code I use is from:
>> http://www.ensembl.org/info/docs/api/core/core_tutorial.html
>>
>> Code:
>> - $slice_adaptor->fetch_by_region( 'chromosome', '11', 123000000,
>> 123000020, 'NCBI36' ); or
>> - $slice_adaptor->fetch_by_region( 'chromosome', '11', 123000000,
>> 123000020, 1, 'NCBI36' );
>>
>> Results when using ensembl website:
>>> 11 dna:chromosome chromosome:NCBI36:11:123000000:123000020:1
>> AAGAGTCCCACCAGCTCCAGG
>>> 11 dna:chromosome chromosome:GRCh37:11:123000000:123000020:1
>> TGCACTCCAGCCTGGGCAATG
>>
>> Thanks,
>> Stef
>> __________________
>> http://www.fastmail.fm
>>
>>
>> _______________________________________________
>> Dev mailing list
>> Dev at ensembl.org
>> http://lists.ensembl.org/mailman/listinfo/dev
>
>
> __________________
> http://www.fastmail.fm
>
>
> _______________________________________________
> Dev mailing list
> Dev at ensembl.org
> http://lists.ensembl.org/mailman/listinfo/dev

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2063 bytes
Desc: not available
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20100916/e754e994/attachment.p7s>