pelikan | 30 Nov 23:19 2006
Picon

Error with supplied lineages importing uniprot data

Hello all,

 I'm running bioperl 1.5.2, bioperl-db 1.5.2 - RC005, under windows,
without Cygwin. The "make test"s have all completed without error. This
is my first time dealing with bioperl, so bear with me.

   I've successfully loaded the most recent taxonomy information using the
biosql-schema scripts. After this, I attempted to load the most recent
release of the uniprot flat file dataset with the following command:

load_seqdatabase.pl -drive mysql -dbname bioseqdb -dbuser root -dbpass
********* -format swiss -safe c:\data\uniprot\uniprot_sprot.dat

I am subsequently greeted by many of the following errors:

Could not store Q7N3Q6:
------------- EXCEPTION: Bio::Root::Exception -------------
MSG: The supplied lineage does not start near 'Photorhabdus luminescens
subsp. laumondii'
STACK: Error::throw
STACK: Bio::Root::Root::throw C:/Perl/site/lib/Bio\Root\Root.pm:359
STACK: Bio::Species::classification C:/Perl/site/lib/Bio\Species.pm:166
STACK: Bio::DB::Persistent::PersistentObject::AUTOLOAD
C:/Perl/site/lib/Bio/DB/Persistent/PersistentObject.pm:55
STACK: Bio::DB::BioSQL::SpeciesAdaptor::populate_from_row
C:/Perl/site/lib/Bio\DB\BioSQL\SpeciesAdaptor.pm:281
STACK: Bio::DB::BioSQL::BasePersistenceAdaptor::_build_object
C:/Perl/site/lib/Bio/DB/BioSQL/BasePersistenceAdap
STACK: Bio::DB::BioSQL::BasePersistenceAdaptor::_find_by_unique_key
C:/Perl/site/lib/Bio/DB/BioSQL/BasePersisten
(Continue reading)

Hilmar Lapp | 1 Dec 00:31 2006
Picon
Picon

Re: Error with supplied lineages importing uniprot data

Maybe Sendu can shed light on what might be happening here?

On Nov 30, 2006, at 5:19 PM, pelikan <at> cs.pitt.edu wrote:

> Hello all,
>
>  I'm running bioperl 1.5.2, bioperl-db 1.5.2 - RC005, under windows,
> without Cygwin. The "make test"s have all completed without error.  
> This
> is my first time dealing with bioperl, so bear with me.
>
>    I've successfully loaded the most recent taxonomy information  
> using the
> biosql-schema scripts. After this, I attempted to load the most recent
> release of the uniprot flat file dataset with the following command:
>
> load_seqdatabase.pl -drive mysql -dbname bioseqdb -dbuser root -dbpass
> ********* -format swiss -safe c:\data\uniprot\uniprot_sprot.dat
>
> I am subsequently greeted by many of the following errors:
>
> Could not store Q7N3Q6:
> ------------- EXCEPTION: Bio::Root::Exception -------------
> MSG: The supplied lineage does not start near 'Photorhabdus  
> luminescens
> subsp. laumondii'
> STACK: Error::throw
> STACK: Bio::Root::Root::throw C:/Perl/site/lib/Bio\Root\Root.pm:359
> STACK: Bio::Species::classification C:/Perl/site/lib/Bio\Species.pm: 
> 166
(Continue reading)

Sendu Bala | 1 Dec 10:03 2006
Picon

Re: Error with supplied lineages importing uniprot data

pelikan <at> cs.pitt.edu wrote:
> Hello all,
> 
>  I'm running bioperl 1.5.2, bioperl-db 1.5.2 - RC005, under windows,
> without Cygwin. The "make test"s have all completed without error. This
> is my first time dealing with bioperl, so bear with me.
> 
>    I've successfully loaded the most recent taxonomy information using the
> biosql-schema scripts. After this, I attempted to load the most recent
> release of the uniprot flat file dataset with the following command:
> 
> load_seqdatabase.pl -drive mysql -dbname bioseqdb -dbuser root -dbpass
> ********* -format swiss -safe c:\data\uniprot\uniprot_sprot.dat
> 
> I am subsequently greeted by many of the following errors:
> 
> Could not store Q7N3Q6:
> ------------- EXCEPTION: Bio::Root::Exception -------------
> MSG: The supplied lineage does not start near 'Photorhabdus luminescens
> subsp. laumondii'

In your uniprot_sprot.dat file there'll be some kind of entry with that 
Photorhabdus species. Can you post that entry (sans sequence if it has 
one) so I can take a look at it? Maybe post a few that cause problems, 
and a few that don't.
Sendu Bala | 1 Dec 18:12 2006
Picon

Re: Error with supplied lineages importing uniprot data

pelikan <at> cs.pitt.edu wrote:
> Hello all,
> 
>  I'm running bioperl 1.5.2, bioperl-db 1.5.2 - RC005, under windows,
> without Cygwin. The "make test"s have all completed without error. This
> is my first time dealing with bioperl, so bear with me.
> 
>    I've successfully loaded the most recent taxonomy information using the
> biosql-schema scripts. After this, I attempted to load the most recent
> release of the uniprot flat file dataset with the following command:
> 
> load_seqdatabase.pl -drive mysql -dbname bioseqdb -dbuser root -dbpass
> ********* -format swiss -safe c:\data\uniprot\uniprot_sprot.dat
> 
> I am subsequently greeted by many of the following errors:
> 
> Could not store Q7N3Q6:

I extracted just Q7N3Q6 from 
ftp://ftp.expasy.org/databases/uniprot/current_release/knowledgebase/complete/uniprot_sprot.dat.gz
and was able to load it in using load_seqdatabase.pl under linux with no 
errors. If you make a file with just that sequence do you still get the 
error?

Is anyone else able to reproduce the problem?
Sendu Bala | 1 Dec 19:52 2006
Picon

Re: Error with supplied lineages importing uniprot data

Sendu Bala wrote:
> pelikan <at> cs.pitt.edu wrote:
[snip]
>> load_seqdatabase.pl -drive mysql -dbname bioseqdb -dbuser root -dbpass
>> ********* -format swiss -safe c:\data\uniprot\uniprot_sprot.dat
>>
>> I am subsequently greeted by many of the following errors:
>>
>> Could not store Q7N3Q6:
> 
> I extracted just Q7N3Q6 from 
> ftp://ftp.expasy.org/databases/uniprot/current_release/knowledgebase/complete/uniprot_sprot.dat.gz
> and was able to load it in using load_seqdatabase.pl under linux with no 
> errors. If you make a file with just that sequence do you still get the 
> error?
> 
> Is anyone else able to reproduce the problem?

In fact, if I just try and load it again I reproduce the problem.
The situation is similar to http://bugzilla.bioperl.org/show_bug.cgi?id=2092

And I have a tentative fix that extends Brian's fix there. Committed to 
HEAD only atm. I don't know anything about bioperl-db and don't have the 
faintest clue why this is happening, nor the time to figure it out. Can 
someone please have a proper look at this and decide if my fix is sane?

All I can say is the the test suites for bioperl-live and bioperl-db 
continue to pass, but that isn't really saying much.

PS. having used load_seqdatabase.pl to load a sequence, how do I remove 
(Continue reading)

Chris Fields | 1 Dec 20:09 2006
Picon

Re: Error with supplied lineages importing uniprot data


On Dec 1, 2006, at 12:52 PM, Sendu Bala wrote:

>
> PS. having used load_seqdatabase.pl to load a sequence, how do I  
> remove
> it afterwards?

There's not much documentation on it, but it demonstrated several  
times in the test suite.

Christopher Fields
Postdoctoral Researcher
Lab of Dr. Robert Switzer
Dept of Biochemistry
University of Illinois Urbana-Champaign
Hilmar Lapp | 2 Dec 02:42 2006
Picon
Picon

Re: Error with supplied lineages importing uniprot data

Either using SQL:

	-- theoretically you should convince yourself first that there
	-- is only one such record (the UK is over acc,version,namespace)
	SQL> DELETE FROM bioentry WHERE accession = 'Q7N3Q6';

or through bioperl-db (see the delete test for examples):

	my $db = Bio::DB::BioDB->new(....);
	my $seq = Bio::PrimarySeq->new(-accession_number=>'Q7N3Q6',
	                               -namespace=>'whatever you used when  
loading');
	my $adp = $db->get_persistence_adaptor($seq);
	my $pseq = $adp->find_by_unique_key($seq);
	$pseq->remove();
	$pseq->commit();

-hilmar

On Dec 1, 2006, at 1:52 PM, Sendu Bala wrote:

> PS. having used load_seqdatabase.pl to load a sequence, how do I  
> remove
> it afterwards?

--

-- 
===========================================================
: Hilmar Lapp  -:-  Durham, NC  -:-  hlapp at gmx dot net :
===========================================================
(Continue reading)

Chris Fields | 1 Dec 20:00 2006
Picon

Re: Error with supplied lineages importing uniprot data


On Dec 1, 2006, at 11:12 AM, Sendu Bala wrote:

> pelikan <at> cs.pitt.edu wrote:
>> Hello all,
>>
>>  I'm running bioperl 1.5.2, bioperl-db 1.5.2 - RC005, under windows,
>> without Cygwin. The "make test"s have all completed without error.  
>> This
>> is my first time dealing with bioperl, so bear with me.
>>
>>    I've successfully loaded the most recent taxonomy information  
>> using the
>> biosql-schema scripts. After this, I attempted to load the most  
>> recent
>> release of the uniprot flat file dataset with the following command:
>>
>> load_seqdatabase.pl -drive mysql -dbname bioseqdb -dbuser root - 
>> dbpass
>> ********* -format swiss -safe c:\data\uniprot\uniprot_sprot.dat
>>
>> I am subsequently greeted by many of the following errors:
>>
>> Could not store Q7N3Q6:
>
> I extracted just Q7N3Q6 from
> ftp://ftp.expasy.org/databases/uniprot/current_release/ 
> knowledgebase/complete/uniprot_sprot.dat.gz
> and was able to load it in using load_seqdatabase.pl under linux  
> with no
(Continue reading)

Chris Fields | 1 Dec 20:06 2006
Picon

Re: Error with supplied lineages importing uniprot data


On Dec 1, 2006, at 11:12 AM, Sendu Bala wrote:

> pelikan <at> cs.pitt.edu wrote:
>> Hello all,
>>
>>  I'm running bioperl 1.5.2, bioperl-db 1.5.2 - RC005, under windows,
>> without Cygwin. The "make test"s have all completed without error.  
>> This
>> is my first time dealing with bioperl, so bear with me.
>>
>>    I've successfully loaded the most recent taxonomy information  
>> using the
>> biosql-schema scripts. After this, I attempted to load the most  
>> recent
>> release of the uniprot flat file dataset with the following command:
>>
>> load_seqdatabase.pl -drive mysql -dbname bioseqdb -dbuser root - 
>> dbpass
>> ********* -format swiss -safe c:\data\uniprot\uniprot_sprot.dat
>>
>> I am subsequently greeted by many of the following errors:
>>
>> Could not store Q7N3Q6:
>
> I extracted just Q7N3Q6 from
> ftp://ftp.expasy.org/databases/uniprot/current_release/ 
> knowledgebase/complete/uniprot_sprot.dat.gz
> and was able to load it in using load_seqdatabase.pl under linux  
> with no
(Continue reading)

Sendu Bala | 1 Dec 20:39 2006
Picon

Re: Error with supplied lineages importing uniprot data

Chris Fields wrote:
> 
> On Dec 1, 2006, at 11:12 AM, Sendu Bala wrote:
> 
>> pelikan <at> cs.pitt.edu wrote:
>>> Hello all,
>>>
>>>  I'm running bioperl 1.5.2, bioperl-db 1.5.2 - RC005, under windows,
>>> without Cygwin. The "make test"s have all completed without error. This
>>> is my first time dealing with bioperl, so bear with me.
>>>
>>>    I've successfully loaded the most recent taxonomy information 
>>> using the
>>> biosql-schema scripts. After this, I attempted to load the most recent
>>> release of the uniprot flat file dataset with the following command:
>>>
>>> load_seqdatabase.pl -drive mysql -dbname bioseqdb -dbuser root -dbpass
>>> ********* -format swiss -safe c:\data\uniprot\uniprot_sprot.dat
>>>
>>> I am subsequently greeted by many of the following errors:
>>>
>>> Could not store Q7N3Q6:
>>
>> I extracted just Q7N3Q6 from
>>
ftp://ftp.expasy.org/databases/uniprot/current_release/knowledgebase/complete/uniprot_sprot.dat.gz 
>>
>> and was able to load it in using load_seqdatabase.pl under linux with no
>> errors. If you make a file with just that sequence do you still get the
>> error?
(Continue reading)

Chris Fields | 1 Dec 20:52 2006
Picon

Re: Error with supplied lineages importing uniprot data

> > 
> > Okay, just updated to get your latest CVS fixes for 
> bioperl-live and 
> > it passes now for both Mac OS X and WinXP.
> 
> Can you confirm if it is actually working correctly though? 
> Like, having stored a previously-problem sequence, can you 
> get it back out from the database and is its ->species() correct?

I would assume so, if we can trust the species tests.  I will have to try it
again over the weekend.  I planned on loading a ton of protein sequences in
anyway, most of which are bacterial; if anything breaks it will probably be
with those.

I think Jason and Hilmar were going to get together about the BioSQL paper
at the hackathon.  That may be a good place to bring some of the species
issues, if they persist.

chris

Gmane