[ensembl-dev] Inquiry: Cafe.v3 Input

Matthieu Muffato muffato at ebi.ac.uk
Tue Jul 30 09:27:29 BST 2013


Hi Mike

Just a few more comments after Magali's email.
The variable $human_gene_adaptor is actually not used in the script. You 
can also just comment out that line. The "cat" parameter is read from 
the taxon_id you give on the command line

You may also want to update:
  79         my $gene = $member->get_Gene;
and replace it with
  79         my $gene = $member->gene_member->get_Gene;
to get rid of a deprecation warning

Best,
Matthieu

On 30/07/13 09:12, Magali wrote:
> Hi Mike,
>
> Our live mirrors only hold the last two releases.
> Giving we are on release 72, you will find 71 and 72 databases there,
> but not 70, which is the db_version you specified for your registry.
>
> As for the species, if you want to use cat data, you probably want to
> replace the human gene_adaptor with a cat one.
>
>     L33: my $human_gene_adaptor = $reg->get_adaptor("Homo sapiens",
>     "core", "Gene");
>
> with
>
>     L33: my $cat_gene_adaptor = $reg->get_adaptor("Felis catus", "core",
>     "Gene");
>
>
>
> Hope that helps,
> Magali
>
>
> On 29/07/13 23:16, Mike Montague wrote:
>> Hello,
>>
>> I submitted the inquiry (below) to Miguel Pignatelli and he suggested
>> that I forward the question to you. Upon looking at our error
>> messages, he suspects we're having an issue connecting with the
>> database or else it may be a general API issue. I've attached the
>> script that we are attempting to use.
>>
>> Thank you for your assistance -
>> Mike Montague
>> ____________
>>
>>
>> Michael J. Montague, Ph.D.
>> Post-Doctoral Research Associate
>> The Genome Institute, Washington University School of Medicine
>> 4444 Forest Park Avenue, Campus Box 8501
>> St. Louis, MO 63108
>> T. 718.809.4093
>>
>> Hello Miguel,
>>
>> With the release of Cafe.v3, we intend to use your script to obtain
>> data from Ensembl for input into Cafe. We are interested in
>> gene expansion outliers for the cat lineage (ID 9685) relative
>> to Carnivora (ID 33554).
>>
>> We made a copy of get_expansions.pl (attached below) and added line 10
>> to call on the Ensembl API here at the Genome Institute. It seems to
>> load fine (line 28).
>>
>>     L10: use lib '/gscmnt/sata206/techd/ensembl_api/v70/ensembl/modules';
>>
>>
>> The human adapter loads fine (line 33) but we're wondering what it has
>> to do with cat?
>>
>>     L33: my $human_gene_adaptor = $reg->get_adaptor("Homo sapiens",
>>     "core", "Gene");
>>
>>
>> The problem is with line 35, which returns nothing so it crashes in
>> lines 38, 39 and 40.
>>
>>     L35: my $comparaDBA =
>>     Bio::EnsEMBL::Registry->get_DBAdaptor('Multi', 'compara');
>>     L38: my $gene_member_adaptor   = $comparaDBA->get_GeneMemberAdaptor;
>>
>>
>> Error:
>>
>>
>>     test_get_expansions.pl 9685 > cat_expansions.txt
>>
>>     Can't call method "get_GeneMemberAdaptor" on an undefined value at
>>     /gscuser/mmontagu/bin/test_get_expansions.pl line 38.
>>     cafe_output$
>>
>> or
>>
>>     perl ~/bin/test_get_expansions.pl 9685 > cat_expansions.txt
>>
>>     DBD::mysql::db selectall_arrayref failed: Can't read dir of '.'
>>     (errno: 24) at
>>     /gscmnt/sata206/techd/ensembl_api/v70/ensembl/modules/Bio/EnsEMBL/Registry.pm
>>     line 1600.
>>
>>     -------------------- WARNING ----------------------
>>     MSG: Homo sapiens is not a valid species name (check DB and API
>>     version)
>>     FILE: Bio/EnsEMBL/Registry.pm LINE: 1187
>>     CALLED BY: Bio/EnsEMBL/Registry.pm  LINE: 972
>>     Date (localtime)    = Wed Jul 24 14:14:00 2013
>>     Ensembl API version = 70
>>     ---------------------------------------------------
>>
>>     -------------------- EXCEPTION --------------------
>>     MSG: Can not find internal name for species 'Homo sapiens'
>>     STACK Bio::EnsEMBL::Registry::get_adaptor
>>     /gscmnt/sata206/techd/ensembl_api/v70/ensembl/modules/Bio/EnsEMBL/Registry.pm:974
>>     STACK toplevel /gscuser/mmontagu/bin/test_get_expansions.pl:33
>>     Date (localtime)    = Wed Jul 24 14:14:00 2013
>>     Ensembl API version = 70
>>     ---------------------------------------------------
>>
>>
>> Mainly, we were wondering if the inputs for lines 33 and 35 are
>> correct for cat?
>>
>>
>> Or alternatively, are we calling on an incorrect version of the API?
>> Or, perhaps were overlooking something else altogether?
>>
>>
>>
>>
>>> *From: *mp at ebi.ac.uk <mailto:mp at ebi.ac.uk>
>>> *Subject: **Re: Inquiry: Cafe.v3 Input*
>>> *Date: *July 27, 2013 9:28:02 AM CDT
>>> *To: *"Mike Montague" <mmontagu at genome.wustl.edu
>>> <mailto:mmontagu at genome.wustl.edu>>
>>> *Cc: *mp at ebi.ac.uk <mailto:mp at ebi.ac.uk>, "Wes Warren"
>>> <wwarren at genome.wustl.edu <mailto:wwarren at genome.wustl.edu>>
>>
>>> Dear Mike,
>>>
>>> I am out of the office for some weeks with very limited access to the
>>> internet at the moment. Looking at your error, it looks like the problem
>>> is trying to connect with the DB or a general ensembl API problem. I
>>> don't
>>> have at hand the script, but if you can send the interesting parts to
>>> ensembl-dev at ebi.ac.uk <mailto:ensembl-dev at ebi.ac.uk> you may be able
>>> to find a useful answer in short
>>> time.
>>> I will take a look as soon as I can once I am back.
>>>
>>> Cheers,
>>>
>>> M;
>>>
>>>
>>> *From: *Wes Warren <wwarren at genome.wustl.edu
>>> <mailto:wwarren at genome.wustl.edu>>
>>> *Subject: **Fwd: Re: Fwd: dN/dS*
>>> *Date: *July 9, 2013 1:48:47 PM CDT
>>> *To: *Mike Montague <mmontagu at genome.wustl.edu
>>> <mailto:mmontagu at genome.wustl.edu>>
>>>
>>> Mike,
>>>
>>> The scripts you will need to retrieve cat data for cafe3. For cafe3
>>> you need 1) a data file containing
>>> gene family sizes for the taxa included in the tree and 2) a Newick
>>> formatted tree, including branch lengths for each gene family.
>>>
>>> Wes
>>>
>>> -------- Original Message -------
>>> Subject: 	Re: Fwd: dN/dS
>>> Date: 	Fri, 08 Mar 2013 17:24:06 +0000
>>> From: 	Miguel Pignatelli <mp at ebi.ac.uk>
>>> To: 	Wes Warren <wwarren at genome.wustl.edu>
>>> CC: 	Javier Herrero <jherrero at ebi.ac.uk>, Matthieu Muffato
>>> <muffato at ebi.ac.uk>
>>>
>>>
>>>
>>> Hi Wes,
>>>
>>> The CAFE data is stored in the compara database and is accessible via
>>> the compara API.
>>>
>>> I am sending you a file with the 43 genes expanded in cat with respect
>>> to the carnivora taxon.
>>>
>>> The columns in the file are:
>>>
>>> gene_tree_root_id, gene names, number of members in cat, number of
>>> members in the parent (carnivora), newick format of the species tree.
>>>
>>> A couple of comments about this data:
>>>
>>> - The newick tree contains the species names and the taxon_ids for the
>>> internal nodes. The numbers are the number of members for each taxon.
>>> - I have only considered significant expansions as reported by our CAFE
>>> analysis. It would be possible to include genes in the "grey" zone (i.e.
>>> genes with more members in cat vs carnivora, but the change not being
>>> significant).
>>>
>>> I am also sending you the script I have written to retrieve the data. It
>>> accepts a taxon_id as input. I called it as follows:
>>>
>>> $ perl get_expansions.pl 9685 > cat_expansions.txt
>>>
>>> If you run the script with other taxon_ids, be aware that it also
>>> reports genes for which the lowest common ancestor is the taxon_id of
>>> interest (i.e. births at the taxon_id of interest).
>>>
>>> Let me know if you have any doubt or need further help in getting the
>>> data you want.
>>>
>>> Cheers,
>>>
>>> M;
>>>
>>>
>>> On 07/03/13 20:15, Wes Warren wrote:
>>> > Hi Miguel,
>>> >
>>> > Is this data available for cat? All the gene trees are available so I
>>> > assume so but I am not sure where to find the CAFE files.
>>> >
>>> > Thanks,
>>> > Wes
>>> >
>>> > On 3/7/13 12:15 PM, Javier Herrero wrote:
>>> >> Hi Wes
>>> >>
>>> >> This is a question for Miguel, our CAFE expert.
>>> >>
>>> >> Javier
>>> >>
>>> >> On 06/03/13 17:00, Wes Warren wrote:
>>> >>> Is the CAFE output available for cat? We are interested in gene
>>> >>> expansion outliers for the cat lineage, perhaps correlating these
>>> >>> with carnivora.
>>> >>>
>>> >>> Wes
>>> >>>
>>




More information about the Dev mailing list