[Bioperl-l] Error while running load_seqdatabase.pl
George Heller
george.heller at yahoo.com
Wed Jan 31 23:15:25 UTC 2007
Hi Hilmar,
I am extremely sorry about sending such a huge file, I realized only after I sent it. Really sorry....
And, finally, the loading worked. I set the primary_id, accession and the display_id to the first number on my file, and it loads perfectly fine now!
Thank you so much Hilmar for all the help.
George.
Hilmar Lapp <hlapp at gmx.net> wrote:
Hi George,
your reply to mine contained a plain-text file attachment of (roughly) 8MB. Not only does it hang up my email reader (probably because it recognizes the files as text and tries to display it), but you also sent it to the entire list, resulting in 8MB being sent to around 1600 people.
Please, whenever you have larger files, only send it to an individual, *never* to a mailing list unless specifically asked to do so.
Second, whenever you have large text files to attach, *always* compress them. This will typically save 70% of the bandwidth being used for transmission.
What I could recover from your file looked like the first element is unique. It also looks like the last 4 elements are the accession of the target sequence (presumably of a GenBank entry - presumably maize BAC clones?) and the start and end coordinates and the strand.
All of display_id, accession_number, and primary_id must be 40 chars or less (sorry for not pointing that out right away), so you may use the first element for all three of them. You could also use some combination of the other elements to construct a more meaningful display_id, just don't exceed 40 chars.
Setting primary_id() to undef should definitely work. If it doesn't you may be using an old version of Bioperl. You should be using release 1.5.2, or at least 1.5.x. It's possible that the 1.4 release still had the bug of not allowing the setting to undef, please upgrade if that's the version you're using.
Let me know if you still have problems with getting this to work.
-hilmar
On Jan 30, 2007, at 12:32 PM, George Heller wrote:
Hi Hilmar,
I have been trying to get around this problem for some days now, but havent had much luck. My file has about 12000 odd records, and when I try to load it with a pipeline to the package, I have about 483 records that get loaded.
sub process_seq
{
my ($self, $seq) = @_;
# $seq->accession_number($seq->display_id);
my @ids = split(/\|/,$seq->display_id);
$seq->accession_number($ids[0]);
$seq->primary_id($ids[2]);
return ($seq);
}
I am assuming this is because I use the ids[2] for the primary_id, and possibly the file has 483 unique records for that field. When I use some other reference like ids[3] etc(coz it has more unique values), I get the same error about the length being more than 40.
I printed out the values for the display_id, primary_id and accession_number. The display_id has the entire first line of the file, so does the primary_id, if I dont set it in my new script. The accession number is split correctly and printed.
This is how the first few lines in my file look like,
>FGENESHT0000001||AC155633|570|4400|1
MTERKRKEIEDRKRKISGPQPGSSNRPRFSGNQPQQFRQNQRPPQQHQQFQRQYPQHQYQNRQSNQSGGQFQRQNQQAPR
LPAPAAQQNSQATPAQVGNRACFHCGEQGHWVMQCPKKAAQQQSGPNAPAKQNVPQPRAGNRSQPRYNHGRLNHLEAEAV
QETPSMIVGMFPVDSHIAEVLFDTGATHSFITASWVEAHNLPITTMSTPIQIDSAGGRIRADSICLNICVEIRGIAFPAN
LIVMGTQGIDVILGMNWLDKYQAVISCDKRTIKLMSPLGEEVVTELVPPEPKRGSCYQLAVDSSEVDPIESIRVVSEFPD
VFPKDLPGMPPERKVEFAIELLPGTAPIFKRAYRISGPELVELKEQIDELSEKGYIRPSTSPWAAPVLFVEKKDGTKRMC
IDYRALNEVTIKNKYPLPRIEDLFDQLRGASVFSKIDLRSAFFMNLMNSVFMDYLDKFVVVFIDDILVYSQSEEEHADHL
KMVLQRLREHQLYAKLSKCEFWINEVLFLGHIINKEGLAVDPKKVANILNWKAPTDARGIKSFIGMVGYYRRFIEGFSKI
AKPMTALLGNKVEFKWTQKCQEAFEALKEKLTIAPVLVLPDVHKPFSVYCDACYTGLGCVLMQEGRVVAYSSRQLKVHEK
NYPIHDLELAAVVHALKTWRHYLYGQKCDVYTDHKSLKYIFTQSELNMRQRRWLELIKDYELEIHYHPGKANVVADALSR
KSQVNLMVARPMPYELAKEFDRLSLGFLNNSRGVTVELEPTLEREIKEAQKNDEKISEIRRLILDGRGKDFREDAEGVIW
FKDRLCVPNVQSIRELILKEAHETAYSIHPGSEKMYQDLKKKFWWYGMKREIAEHVAMCDSCRRIKAEHQRPAGLLQPLQ
IPQWKWDEIGMDFI VGLPRTRAGYDSIWVVVDRLTKSAHFIPVKTNYSSAVLAELYMSRIVCLHGVPKKIVSDRGTQFTS
HFWRQLHEALGTHLNFSSAYHPQTDGQTERTNQILEDMLRACALQDQSGWDKRLPYAEFSYNNSYQASLKMSPFQALYGR
SCRTPLQWDQPGEKQVFGPDILLEAEENIKMVRENLKIAQSRQRSYADTRRRELSFEVGDFVYLKVSPIRGVKRFGVKGK
LAPRYIGSYQILARRGEVAYQLSLPENLSAVHDVFHVSQLKKCLRVPEEQLPVEGLEVQEDLTYVEKPVQILEVADRVTR
RKTIRMCKVRWNHHSEEEATSEREDDLMAKYPELFASQP*
Any suggestions?
Thanks!
George.
Hilmar Lapp <hlapp at gmx.net> wrote:
That's odd indeed. Did you try and put a print statement before the
return statement that proves that 1) the codes gets executed, and 2)
display_id(), primary_id() and accession_number() have the expected
values?
BTW you might also want to set primary_id() to undef (as the ID found
in the FASTA files doesn't really count as a primary database-
specific ID anyway). The identifier column in bioentry (which
primary_id() maps to) is constrained to 40 chars as well.
-hilmar
On Jan 27, 2007, at 9:31 PM, George Heller wrote:
> Hi Hilmar,
>
> I tried the lookup and noupdate options, and also made changes to
> the SeqProcessor.pm package for the accession,
> package SeqProcessor::Accession;
> use strict;
> use vars qw(@ISA);
> use Bio::Seq::BaseSeqProcessor;
> use Bio::SeqFeature::Generic;
> @ISA = qw(Bio::Seq::BaseSeqProcessor);
> sub process_seq
> {
> my ($self, $seq) = @_;
> my @ids = split(/|/,$seq->display_id);
> $seq->accession_number($ids[0]);
> return ($seq);
> }
> 1;
> I invoke the load_seqdatabase.pl as,
> perl load_seqdatabase.pl -host localhost -dbname usda-06 -format
> fasta -dbuser postgres -driver Pg --lookup --noupdate --
> pipeline="SeqProcessor::Accession" maize_pep.fasta
> Loading maize_pep.fasta ...
> I get the error,
> -------------------- WARNING ---------------------
> MSG: insert in Bio::DB::BioSQL::SeqAdaptor (driver) failed, values
> were ("FGENESHT0000021||AC155633|113788|
> 114708|-1","FGENESHT0000021||AC155633|113788|
> 114708|-1","FGENESHT0000021||AC155633|113788|114708|-1","","0","")
> FKs (1,)
> ERROR: value too long for type character varying(40)
> ---------------------------------------------------
> Could not store FGENESHT0000021||AC155633|113788|114708|-1:
> ------------- EXCEPTION -------------
> MSG: error while executing statement in
> Bio::DB::BioSQL::SeqAdaptor::find_by_unique_key: ERROR: current
> transaction is aborted, commands ignored until end of transaction
> block
> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::_find_by_unique_key /
> home/akar/local/perl//Bio/DB/BioSQL/BasePersistenceAdaptor.pm:951
> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::find_by_unique_key /
> home/akar/local/perl//Bio/DB/BioSQL/BasePersistenceAdaptor.pm:855
> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::create /home/akar/
> local/perl//Bio/DB/BioSQL/BasePersistenceAdaptor.pm:205
> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::store /home/akar/
> local/perl//Bio/DB/BioSQL/BasePersistenceAdaptor.pm:254
> STACK Bio::DB::Persistent::PersistentObject::store /home/akar/local/
> perl//Bio/DB/Persistent/PersistentObject.pm:272
> STACK (eval) load_seqdatabase.pl:620
> STACK toplevel load_seqdatabase.pl:602
> --------------------------------------
> at load_seqdatabase.pl line 633
> As far as I gather, this error shouldnt appear as we are
> filtering out the accession as only the first code that appears.
> Ideas?
>
> George.
>
>
> Hilmar Lapp wrote:
> George,
>
> I don't know you create the FASTA file, but that's probably where the
> root cause is. Based on the message:
>
>> -------------------- WARNING ---------------------
>> MSG: insert in Bio::DB::BioSQL::SeqAdaptor (driver) failed, values
>> were ("FGENESHT0000001||AC155633|570|4400|1","FGENESHT0000001||
>> AC155633|570|4400|1","FGENESHT0000001||AC155633|570|4400|
>> 1","","0","") FKs (1,)
>> ERROR: duplicate key violates unique constraint
>> "bioentry_accession_key"
>> ---------------------------------------------------
>
> the identifier and accession number are set, so your SeqProcessor
> scriptlet was executed (otherwise you'd also have seen a dynamic
> loading error it e.g. you perl class could be not be found or loaded
> by perl). If you still receive the duplicate key violation, then it
> can only mean that indeed a sequence with the exact same accession
> number was in the database already.
>
> There are different possibilities for why: you may have loaded the
> same file before (use --lookup and related switches if you want to
> update existing sequences), or your FASTA file contains multiple
> sequences with the same ID, or you have a sequence with the same ID
> in different FASTA files, if you are loading from more than one file.
> In either of the two latter cases, you will need to find a way to
> disambiguate the IDs.
>
> BTW you also want to consider to parse the concatenated ID
> 'FGENESHT0000001||AC155633|570|4400|1' apart into its component IDs,
> and then use only one component. For example:
>
> my @ids = split(/|/,$seq->display_id);
> $seq->accession_number($ids[0]);
>
> Obviously, this will only make for a nicer accession number, and not
> solve your duplicate ID problem, as the latter is in the file(s) you
> load.
>
> -hilmar
>
> On Jan 25, 2007, at 8:51 PM, George Heller wrote:
>
>> Hi Hilmar,
>>
>> I still seem to be having problems loading my fasta file. I wrote a
>> new package, SeqProcessor.pm as below,
>>
>> package SeqProcessor::Accession;
>> use strict;
>> use vars qw(@ISA);
>> use Bio::Seq::BaseSeqProcessor;
>> use Bio::SeqFeature::Generic;
>> @ISA = qw(Bio::Seq::BaseSeqProcessor);
>> sub process_seq
>> {
>> my ($self, $seq) = @_;
>> $seq->accession_number($seq->display_id);
>> return ($seq);
>> }
>> 1;
>> I have this file SeqProcessor.pm in my home directory, and I have
>> set the PERL5LIB variable accordingly. When I run
>> load_seqdatabase.pl,
>>
>> perl load_seqdatabase.pl -host localhost -dbname biodb -format
>> fasta -dbuser postgres -driver Pg --
>> pipeline="SeqProcessor::Accession" maize_pep.fasta
>>
>> I still get the error,
>>
>> Loading maize_pep.fasta ...
>> -------------------- WARNING ---------------------
>> MSG: insert in Bio::DB::BioSQL::SeqAdaptor (driver) failed, values
>> were ("FGENESHT0000001||AC155633|570|4400|1","FGENESHT0000001||
>> AC155633|570|4400|1","FGENESHT0000001||AC155633|570|4400|
>> 1","","0","") FKs (1,)
>> ERROR: duplicate key violates unique constraint
>> "bioentry_accession_key"
>> ---------------------------------------------------
>> Could not store FGENESHT0000001||AC155633|570|4400|1:
>> ------------- EXCEPTION -------------
>> MSG: error while executing statement in
>> Bio::DB::BioSQL::SeqAdaptor::find_by_unique_key: ERROR: current
>> transaction is aborted, commands ignored until end of transaction
>> block
>> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::_find_by_unique_key /
>> home/akar/local/perl//Bio/DB/BioSQL/BasePersistenceAdaptor.pm:951
>> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::find_by_unique_key /
>> home/akar/local/perl//Bio/DB/BioSQL/BasePersistenceAdaptor.pm:855
>> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::create /home/akar/
>> local/perl//Bio/DB/BioSQL/BasePersistenceAdaptor.pm:205
>> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::store /home/akar/
>> local/perl//Bio/DB/BioSQL/BasePersistenceAdaptor.pm:254
>> STACK Bio::DB::Persistent::PersistentObject::store /home/akar/local/
>> perl//Bio/DB/Persistent/PersistentObject.pm:272
>> STACK (eval) load_seqdatabase.pl:620
>> STACK toplevel load_seqdatabase.pl:602
>> --------------------------------------
>> at load_seqdatabase.pl line 633
>> Is there something I am missing?
>>
>> Thanks!
>> George.
>>
>>
>> Hilmar Lapp wrote:
>> Hi George, sorry for the sluggish response, I was tied up during the
>> week. This is also why you always want to keep the thread on the
>> list.
>>
>> Perl is an interpreted language, so no compilation is necessary. The
>> only thing you need to do is have the package in a place where perl
>> can find it. The simplest way to achieve this is by setting the
>> PERL5LIB environment variable:
>>
>> $ export PERL5LIB=/where/you/put/your/perl/package
>>
>> or if PERL5LIB was set already, you'd append it:
>>
>> $ export PERL5LIB=${PERL5LIB}:/where/you/put/your/perl/package
>>
>> I do assume that you didn't really add your code to the SeqAdaptor.pm
>> package - there is no necessity for nor benefit from that, and at
>> worst (and quite likely) perl won't be able to find the package. Note
>> that there is plenty of documentation for how to write packages for
>> perl and how to make them accessible to perl.
>>
>> Hth,
>>
>> -hilmar
>>
>> On Jan 8, 2007, at 11:52 PM, George Heller wrote:
>>
>>> Hi Hilmer.
>>>
>>> Thanks so much for the response. As I am new to Bioperl, I have
>>> another question.
>>>
>>> I have made the changes as suggested by you, and have added the
>>> code below to the SeqAdaptor.pm script.
>>>
>>> package SeqProcessor::Accession;
>>> use strict;
>>> use vars qw(@ISA);
>>> use Bio::Seq::BaseSeqProcessor;
>>> use Bio::SeqFeature::Generic;
>>>
>>> @ISA = qw(Bio::Seq::BaseSeqProcessor);
>>>
>>> sub process_seq
>>> {
>>> my ($self, $seq) = @_;
>>> $seq->accession_number($seq->display_id);
>>> return ($seq);
>>> }
>>>
>>> Now that I have done my changes, do I need to compile or something
>>> for the changes to reflect? If so, can you please let me know the
>>> command for the same, or direct me to any lin that has
>>> documentation for the same?
>>>
>>> Thanks so much for the help.
>>> George.
>>>
>>> Hilmar Lapp wrote:
>>> George,
>>>
>>> this is almost certainly caused by using FASTA format and bioperl's
>>> treatment of it. I am guilty of not having written a FAQ yet for
>>> Bioperl-db, as this would certainly be there.
>>>
>>> Specifically, the Bioperl fasta SeqIO parser (load_seqdatabase.pl
>>> uses Bioperl to parse sequence files) does not extract the accession
>>> number from the description line of the fasta sequence, and instead
>>> sets the accession_number property if sequence objects it creates to
>>> "unknown". Since there is a unique key constraint on
>>> (accession,version,namespace) the second sequence loaded will raise
>>> an exception as it will violate the constraint.
>>>
>>> The simplest way to deal with this is to write a SeqProcessor that
>>> massages the accession_number appropriately and then supply the
>>> module to load_seqdatabase.pl using the --pipeline command line
>>> switch.
>>>
>>> There are several examples for how to do this in the email archives.
>>> See for example this thread on the Biosql list:
>>>
>>> http://lists.open-bio.org/pipermail/biosql-l/2005-August/000901.html
>>>
>>> with two links to examples, and Marc Logghe gives another one in the
>>> thread itself.
>>>
>>> Hth,
>>>
>>> -hilmar
>>>
>>> On Jan 8, 2007, at 3:17 PM, George Heller wrote:
>>>
>>>> Hi all.
>>>>
>>>> I am new to Bioperl and am trying to run the load_seqdatabase.pl
>>>> script to load sequence data from a file into Postgres database. I
>>>> am invoking the script through the following command:
>>>>
>>>> perl load_seqdatabase.pl -host localhost -dbname biodb06 -format
>>>> fasta
>>>> -dbuser postgres -driver Pg
>>>>
>>>> I am getting the following error:
>>>>
>>>> -------------------- WARNING ---------------------
>>>> MSG: insert in Bio::DB::BioSQL::SeqAdaptor (driver) failed, values
>>>> were ("FGENES
>>>> HT0000001||AC155633|570|4400|1","FGENESHT0000001||AC155633|570|
>> 4400|
>>>> 1","unknown"
>>>> ,"","0","") FKs (1,)
>>>> ERROR: duplicate key violates unique constraint
>>>> "bioentry_accession_key"
>>>> ---------------------------------------------------
>>>> Could not store unknown:
>>>> ------------- EXCEPTION -------------
>>>> MSG: error while executing statement in
>>>> Bio::DB::BioSQL::SeqAdaptor::find_by_uni
>>>> que_key: ERROR: current transaction is aborted, commands ignored
>>>> until end of t
>>>> ransaction block
>>>> STACK
>>>> Bio::DB::BioSQL::BasePersistenceAdaptor::_find_by_unique_key /usr/
>>>> lib/perl
>>>> 5/site_perl/5.8.5/Bio/DB/BioSQL/BasePersistenceAdaptor.pm:948
>>>> STACK
>> Bio::DB::BioSQL::BasePersistenceAdaptor::find_by_unique_key /
>>>> usr/lib/perl5
>>>> /site_perl/5.8.5/Bio/DB/BioSQL/BasePersistenceAdaptor.pm:852
>>>> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::create /usr/lib/
>>>> perl5/site_perl/5
>>>> .8.5/Bio/DB/BioSQL/BasePersistenceAdaptor.pm:203
>>>> STACK Bio::DB::BioSQL::BasePersistenceAdaptor::store /usr/lib/
>> perl5/
>>>> site_perl/5.
>>>> 8.5/Bio/DB/BioSQL/BasePersistenceAdaptor.pm:251
>>>> STACK Bio::DB::Persistent::PersistentObject::store /usr/lib/perl5/
>>>> site_perl/5.8.
>>>> 5/Bio/DB/Persistent/PersistentObject.pm:271
>>>> STACK (eval) load_seqdatabase.pl:620
>>>> STACK toplevel load_seqdatabase.pl:602
>>>> --------------------------------------
>>>> at load_seqdatabase.pl line 633
>>>>
>>>> Can anyone tell me how I can correct this error and get my script
>>>> running? Thanks!!!
>>>>
>>>> George.
>>>>
>>>>
>>>> __________________________________________________
>>>> Do You Yahoo!?
>>>> Tired of spam? Yahoo! Mail has the best spam protection around
>>>> http://mail.yahoo.com
>>>> _______________________________________________
>>>> Bioperl-l mailing list
>>>> Bioperl-l at lists.open-bio.org
>>>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>>>
>>> --
>>> ===========================================================
>>> : Hilmar Lapp -:- Durham, NC -:- hlapp at gmx dot net :
>>> ===========================================================
>>>
>>>
>>>
>>>
>>>
>>>
>>> __________________________________________________
>>> Do You Yahoo!?
>>> Tired of spam? Yahoo! Mail has the best spam protection around
>>> http://mail.yahoo.com
>>
>> --
>> ===========================================================
>> : Hilmar Lapp -:- Durham, NC -:- hlapp at gmx dot net :
>> ===========================================================
>>
>>
>>
>>
>>
>>
>>
>> Looking for earth-friendly autos?
>> Browse Top Cars by "Green Rating" at Yahoo! Autos' Green Center.
>
> --
> ===========================================================
> : Hilmar Lapp -:- Durham, NC -:- hlapp at gmx dot net :
> ===========================================================
>
>
>
>
>
>
>
>
> ---------------------------------
> Check out the all-new Yahoo! Mail beta - Fire up a more powerful
> email and get things done faster.
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l
--
===========================================================
: Hilmar Lapp -:- Durham, NC -:- hlapp at gmx dot net :
===========================================================
---------------------------------
Don't pick lemons.
See all the new 2007 cars at Yahoo! Autos.
--
===========================================================
: Hilmar Lapp -:- Durham, NC -:- hlapp at gmx dot net :
===========================================================
---------------------------------
Everyone is raving about the all-new Yahoo! Mail beta.
More information about the Bioperl-l
mailing list