Re: [GUSDEV] Problems in GUS

[F I. <fr...@pc...>]

Hi Diogo,

The first issue does not seem like a bug, but the second possibly is.  For the second one, the problem may be more obvious to someone else in the group.  I'd like to encourage other people to give a answer, but here is what I see so far:

The first, InsertBlastSimilarities, is a supported plugin.  I doesn't look like an error, though; if no "subjects" exist in the sequence tables, the error can be resolved by inserting the missing subjects with the InsertNewExternalSequences plugin.

The second issue is with community plugin, OrthologGroupsMCL, and it seems like a soft error.  I'm not sure I understand the issue - I'm guessing there are sequence IDs in the OrthoMCL file that were not loaded into aasequence?  If so, it would make sense for it to fail, but it should do so more graciously.

-Frank

P.S

Full notes from the InsertBlastSimilarities plugin:
"The definition lines of the sequences involved in the BLAST (both query and subject) must begin with the na_sequence_ids of those sequences.  The standard way to acheive that is to first load the sequences into GUS, using the InsertNewExternalSequences plugin, and then to extract them into a file with the dumpSequences FromTable.pl command.  That command places the na_sequence_id of the sequence as the first thing in the definition line.

The plugin assumes that all subject sequences are stored in the same table, and that all query sequences are stored in the same table (but possibly different from the subject sequences).  The subjectTable and queryTable arguments indicate which tables the sequences are in, respectively.  (This is because the Similarity table forms similarities between table/rows in GUS; therefore, the plugin must specify which tables the sequences are in.)"


>> *From: *"Diogo Tschoeke" <di...@gm... <mailto:di...@gm...>>
>> *Date: *January 30, 2008 9:02:22 AM EST
>> *To: *GUSDEV <Gus...@li...
>> <mailto:Gus...@li...>>
>> *Subject: **[GUSDEV] Problems in GUS*
>>
>> Hello all,
>>
>> I work with GUS on Brazil at Instituto Oswaldo Cruz, and have some
>> questions:
>> In plugin InsertBlastSimilarities when I have a query with zero
>> subject, something like that: >ABQ58803.1 (0 subjects)
>> this result is not inserted on GUS, it's correct?
>> Another question is when I use de pluggin: OrthologGroupsMCL
>> i have this error
>> DBD::Pg::st execute failed: ERROR:  null value in column
>> "aa_sequence_id" violates not-null constraint
>> ERROR:
>>
>>  SQL ERROR!! involving
>>
>>     INSERT INTO DoTS.AASequenceSequenceGroup
>>
>> I know wich aa_sequence_id is a PK of the table dots.aasequenceimp,
>> and the pluggin it's not working because the OrthoMCL plugin not found
>> some sequences in GUS, but OrthoMCL file's have this sequences, it's
>> correct?
>>  
>> Somebody help me?
>>
>> Thank's
>>
>> Diogo Tschoeke
>> Oswaldo Cruz Institute
>> Rio de Janeiro, Brazil
>>