[Gmod-help] Re: [Gmod-schema] Apollo and Chado problems

Scott Cain scott at scottcain.net
Tue Dec 15 15:51:46 EST 2009 

Hi J,

I just wanted to let you know that I've been working on this.  There
are actually two bugs that your experience have pointed out.

1. The way that Bio::FeatureIO::gff and Bio::SeqIO::fasta interact
causes the first sequence in a GFF or fasta file to be lost.
  a. A work around for this bug has been committed to
gmod_gff3_preprocessor.pl so that fasta files created by it won't be
affected anymore.
  b. I'm working on actually fixing Bio::FeatureIO::gff (which is
where the real bug is), but boy did we write some arcane code for that
:-)

2. When I modified the bulk loader in the last few weeks (making it
really much better :-) I broke the ability to load a fasta file
separately as though it were a GFF3 file.  Because I think that is
nice functionality to have, I will either restore that functionality,
or add a --fasta option (whichever turns out to be easier).

Scott


On Thu, Dec 10, 2009 at 5:03 PM, J.M.P. Alves <jmalves at vcu.edu> wrote:
> OK, after quite a bit of poking around the code and DB (sorry, I'm new to
> GMOD, Chado AND PostgreSQL...), I seem to have solved my problem with the
> loading of sequences to the database from GFF. At least now GBrowse is
> working perfectly, and Apollo can also load the sequence fine.
>
> Here goes my solution, in case other people out there face the same problem
> -- and please let me know if I'm doing it wrong...
>
> First, the quick answer:
> - after using gmod_gff_preprocessor.pl as described in the Wiki and getting
> two files (one with annotations in GFF, the other with sequence), use 'cat'
> to join them (I added the sorted GFF part first, the sequence part second),
> like:
>
> cat yeast.gff.sorted yeast.gff.sorted.fasta > yeast_all.gff
>
> - upload the resulting joined file, like this:
>
> gmod_bulk_load_gff3.pl --organism yeast --gfffile yeast_all.gff
>
> Maybe --nosplit to the preprocessor would solve this too, but I haven't
> tried.
>
>
>
> Now, as is my bad habit, for the long answer.
>
> I was having trouble loading the sequence when following the instructions
> from http://gmod.org/wiki/Load_GFF_Into_Chado plus what the perldoc from
> gmod_gff_preprocessor.pl told me:
>
> [quote]
> If the GFF3 file contains FASTA sequence at the end, the sequence will be
> placed in a separate file with the extention '.fasta'.  This fasta file can
> be loaded separately after the split and/or
>       sorted GFF3 files are loaded, using the command:
>
>         gmod_bulk_load_gff3.pl -g <fasta file name>
> [/quote]
>
> Well, so that's exactly what I did, and got the ugly errors I pasted in the
> earlier message.
>
> Problem is that the current version of gmod_bulk_load_gff3.pl does NOT seem
> to have a -g option for loading the sequences as described in the
> preprocessor's documentation -- and as I believe GetOptions does, it is
> interpreting -g as short for --gfffile.
>
> Looking at the code, gmod_bulk_load_gff3.pl seems to create a temporary
> table (tmp_gff_load_cache) during loading of the annotation that seems to no
> longer exist (or contain the necessary data, at least) in a second run of
> the program. I think.
>
> Either way, joining the two files output by gmod_gff_preprocessor.pl
> resulted in the sequence finally being uploaded fine.
>
> So the documentation from both gmod_gff_preprocessor.pl and
> http://gmod.org/wiki/Load_GFF_Into_Chado should be changed accordingly when
> possible. Unless I'm the only bone-headed person who's interpreted things
> the way I have. :-)
>
> Cheers
> J
>
> J.M.P. Alves wrote:
>>
>> Thanks. I noticed that when I accessed GBrowse and saw the display.
>>
>> But another problem happened when I tried to load the sequence (file
>> generated by gmod_gff3_preprocessor.pl), as listed below. I had already
>> loaded the file other file generated by the preprocessor, and the only
>> possibly suspicious things in the output where two skipped tables:
>>
>> Skipping cv table since the load file is empty...
>> Loading data into analysis table ...
>> Skipping organism table since the load file is empty...
>>
>> I'm guessing that the organism table warning is not a problem, given I
>> provided the --organism option. Is the second table, cv, the problem here?
>>
>> The sequence file looks almost completely like a regular FASTA file,
>> except for the very top, which looks like:
>>
>> ##FASTA
>> ##FASTA
>>  >chrI
>>
>> CCACACCACACCCACACACCCACACACCACACCACACACCACACCACACCCACACACACACATCCTAACACTACCCTAAC
>>
>> ACAGCCCTAATCTAACCCTGGCCAACCTGTCTCTCAACTTACCCTCCATTACCCTGCCTCCACTCGTTACCCTGTCCCAT
>>
>> This did not happen with the previous version of the schema and GMOD
>> software an general, as far as I can tell (sure did not get the error when
>> loading, can't remember whether I tried to access the sequence from GBrowse
>> at the time of those tests)
>>
>> Thanks for your help
>> J
>>
>> ===================================================
>> gmod_bulk_load_gff3.pl --organism yeast  -g
>> saccharomyces_cerevisiae.gff.sorted.fasta
>> Preparing data for inserting into the mas_chado database
>> (This may take a while ...)
>> No feature found for
>> CCACACCACACCCACACACCCACACACCACACCACACACCACACCACACCCACACACACACATCCTAACACTACCCTAAC,
>> org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 4.
>> No feature found for chrII, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 5.
>> No feature found for chrIII, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 6.
>> No feature found for chrIV, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 7.
>> No feature found for chrV, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 8.
>> No feature found for chrVI, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 9.
>> No feature found for chrVII, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 10.
>> No feature found for chrVIII, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 11.
>> No feature found for chrIX, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 12.
>> No feature found for chrX, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 13.
>> No feature found for chrXI, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 14.
>> No feature found for chrXII, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 15.
>> No feature found for chrXIII, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 16.
>> No feature found for chrXIV, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 17.
>> No feature found for chrXV, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 18.
>> No feature found for chrXVI, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 19.
>> No feature found for chrMito, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 20.
>> No feature found for 2-micron, org_id:10 when trying to add sequence at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 2479, <GEN0> line
>> 21.
>> Skipping feature table since the load file is empty...
>> Skipping featureloc table since the load file is empty...
>> Skipping feature_relationship table since the load file is empty...
>> Skipping featureprop table since the load file is empty...
>> Skipping feature_cvterm table since the load file is empty...
>> Skipping synonym table since the load file is empty...
>> Skipping feature_synonym table since the load file is empty...
>> Skipping dbxref table since the load file is empty...
>> Skipping feature_dbxref table since the load file is empty...
>> Skipping analysisfeature table since the load file is empty...
>> Skipping cvterm table since the load file is empty...
>> Skipping db table since the load file is empty...
>> Skipping cv table since the load file is empty...
>> Skipping analysis table since the load file is empty...
>> Skipping organism table since the load file is empty...
>> Loading sequences (if any) ...
>>
>> Done.
>>
>> While this script has made an effort to optimize the database, you
>> should probably also run VACUUM FULL ANALYZE on the database as well
>> commit ineffective with AutoCommit enabled at
>> /usr/local/share/perl/5.10.0/Bio/GMOD/DB/Adapter.pm line 1349.
>> ===================================================
>>
>>
>>
>>
>> Scott Cain wrote:
>>>
>>> That's not a problem.  That was just introduced to the codebase within
>>> the last few weeks and I need to fix it, but it is calling a commit at
>>> the end of the load when it isn't needed.
>>>
>>> Scott
>>>
>>> On Thu, Dec 3, 2009 at 5:46 PM, J.M.P. Alves <jmalves at vcu.edu> wrote:
>>>>
>>>> Hi,
>>>>
>>>> At the end of loading the GFF file (using gmod_bulk_load_gff3.pl) into
>>>> Chado, I get something I did not get before:
>>>>
>>>> commit ineffective with AutoCommit enabled
>>>>
>>>> Is that just a warning and the data was loaded anyway, or does it mean
>>>> the
>>>> commit failed? If I knew the schema better, I could check some tables,
>>>> but
>>>> there are so many...
>>>>
>>>> Thanks
>>>> J
>>>>
>>>> Scott Cain wrote:
>>>>>
>>>>> Hi J,
>>>>>
>>>>> Yes, getting schema/trunk is sufficient.  Since we moved to svn pretty
>>>>> recently, there are likely to be other holes in documentation where
>>>>> cvs was assumed.
>>>>>
>>>>> As for further questions, it depends on what they are concerning.  If
>>>>> apollo webstart is causing you problems, you can ask about it on the
>>>>> apollo mailing list, GBrowse questions to the gbrowse list, and most
>>>>> other questions would go to the schema mailing list.
>>>>>
>>>>> Scott
>>>>>
>>>>> On Thu, Dec 3, 2009 at 12:15 PM, J.M.P. Alves <jmalves at vcu.edu> wrote:
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> Thanks for the answer, that helped. I was following the instructions
>>>>>> from
>>>>>> http://gmod.org/wiki/Chado_-_Getting_Started but I missed the part
>>>>>> about
>>>>>> ant
>>>>>> (because I was originally not installing from SVN) -- have that
>>>>>> installed
>>>>>> now. I didn't see anything about the checking out of the whole schema
>>>>>> though.
>>>>>>
>>>>>> What I did was this:
>>>>>>
>>>>>> svn co https://gmod.svn.sourceforge.net/svnroot/gmod/schema/trunk
>>>>>>
>>>>>> It does check out GMODTools and other stuff. Is that sufficient? It
>>>>>> did
>>>>>> take
>>>>>> care of all the errors at any rate.
>>>>>>
>>>>>> I'm now loading the ontologies into my database, and will next load
>>>>>> the
>>>>>> S.
>>>>>> cerevisiae GFF into the DB for testing (as described at
>>>>>> http://gmod.org/wiki/Load_GFF_Into_Chado).
>>>>>>
>>>>>> Thanks again
>>>>>> J
>>>>>>
>>>>>> PS: quick question, I'm using the community annotation server (VMware
>>>>>> image
>>>>>> available from http://gmod.org/wiki/Community_Annotation_System) you
>>>>>> have,
>>>>>> and it's mostly working (webstarted Apollo is giving a little
>>>>>> trouble). I
>>>>>> do
>>>>>> have some problems and questions about it though. Which list should I
>>>>>> direct
>>>>>> them to?
>>>>>>
>>>>>>
>>>>>> Scott Cain wrote:
>>>>>>>
>>>>>>> Hi J,
>>>>>>>
>>>>>>> I don't know how I missed a few emails in this thread, but the
>>>>>>> problem
>>>>>>> you are having is because you need to checkout the entire schema
>>>>>>> repository, not just the chado subdirectory.  You must have ant
>>>>>>> installed, and you need to have the GMODTools directory, which is at
>>>>>>> the same level as the chado directory.
>>>>>>>
>>>>>>> Scott
>>>>>>>
>>>>>>>
>>>>>>> On Wed, Dec 2, 2009 at 4:11 PM, J.M.P. Alves <jmalves at vcu.edu> wrote:
>>>>>>>>
>>>>>>>> Hi, Dave
>>>>>>>>
>>>>>>>> Thanks, that did help -- although the two lines you mention were
>>>>>>>> *already*
>>>>>>>> commented out. I also had to comment out the following:
>>>>>>>>
>>>>>>>> if ($ant or !-e "$working_dir/../GMODTools") {
>>>>>>>>  push @exe_files, 'bin/gmod_bulkfiles.pl',
>>>>>>>> 'bin/gmod_gff2biomart5.pl';
>>>>>>>> }
>>>>>>>>
>>>>>>>> These are currently lines 493-5 of Makefile.PL
>>>>>>>>
>>>>>>>> After that, it did finish without problems, apparently.
>>>>>>>>
>>>>>>>> I am currently updating from the gmod-1.0 file found on SourceForge
>>>>>>>> to
>>>>>>>> the
>>>>>>>> SVN code. Do I need to uninstall the old version (how? 'make
>>>>>>>> uninstall'
>>>>>>>> does
>>>>>>>> not do it, just gives a list of things to delete) or is just
>>>>>>>> installing
>>>>>>>> on
>>>>>>>> top of the one I have right now fine?
>>>>>>>>
>>>>>>>> Thanks
>>>>>>>> J
>>>>>>>>
>>>>>>>> Dave Clements, GMOD Help Desk wrote:
>>>>>>>>>
>>>>>>>>> Hi J,
>>>>>>>>>
>>>>>>>>> We generally recommend searching using Nabble:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>  http://www.nabble.com/Generic-Model-Organism-System-Database-f3491.html
>>>>>>>>>
>>>>>>>>> And, using Nabble, I found a thread from 2007
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> (http://old.nabble.com/Blast-input-file-for-CHADO-to9828403.html#a9847174)
>>>>>>>>> where someone had the same problem.  I have no idea why you are
>>>>>>>>> having
>>>>>>>>> the problem, but a workaround is:
>>>>>>>>>
>>>>>>>>>  But, it works after commenting out the following lines in
>>>>>>>>> Makefile.PL.
>>>>>>>>>  #'bin/gmod_bulkfiles.pl',        line no 522
>>>>>>>>>  #'bin/gmod_gff2biomart5.pl',  line no 523
>>>>>>>>>
>>>>>>>>> You don't need the biomart script unless you are using BioMart too.
>>>>>>>>> Does that work for you?
>>>>>>>>>
>>>>>>>>> Dave C.
>>>>>>>>>
>>>>>>>>> On Mon, Nov 30, 2009 at 1:44 PM, J.M.P. Alves <jmalves at vcu.edu>
>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>> Thanks, Scott
>>>>>>>>>>
>>>>>>>>>> I just subscribed to the list. BTW, is there a way to search it? I
>>>>>>>>>> looked there, but the Sourceforge interface is terrible, and the
>>>>>>>>>> search
>>>>>>>>>> box on the top right searches all the site's project names instead
>>>>>>>>>> of
>>>>>>>>>> the list archives.
>>>>>>>>>>
>>>>>>>>>> I ask because, following your advice, I checked out the svn
>>>>>>>>>> version.
>>>>>>>>>> Then I got a warning during perl Makefile.PL that the kit was
>>>>>>>>>> missing
>>>>>>>>>> files, contact author:
>>>>>>>>>>
>>>>>>>>>> ===========================================
>>>>>>>>>> Checking if your kit is complete...
>>>>>>>>>> Warning: the following files are missing in your kit:
>>>>>>>>>>    bin/cxgn/load_cvterms.pl
>>>>>>>>>>    bin/gmod_bulkfiles.pl
>>>>>>>>>>    bin/gmod_gff2biomart5.pl
>>>>>>>>>>    conf/bulkfiles/anogam.xml
>>>>>>>>>>    conf/bulkfiles/blastfiles.xml
>>>>>>>>>>    conf/bulkfiles/bulkfiles_template.xml
>>>>>>>>>>    conf/bulkfiles/chadofeatconv.xml
>>>>>>>>>>    conf/bulkfiles/chadofeatsql.xml
>>>>>>>>>>    conf/bulkfiles/chadogenepagesql.xml
>>>>>>>>>>    conf/bulkfiles/dmelhetfeatconv.xml
>>>>>>>>>>    conf/bulkfiles/dmelr420.xml
>>>>>>>>>>    conf/bulkfiles/dmelr430.xml
>>>>>>>>>>    conf/bulkfiles/dpsebulk-p4.xml
>>>>>>>>>>    conf/bulkfiles/dpsebulk-p5.xml
>>>>>>>>>>    conf/bulkfiles/dpsebulk-r2.xml
>>>>>>>>>>    conf/bulkfiles/drosmelgb.xml
>>>>>>>>>>    conf/bulkfiles/fastawriter.xml
>>>>>>>>>>    conf/bulkfiles/fbbulk-hetr3.xml
>>>>>>>>>>    conf/bulkfiles/fbbulk-r3.xml
>>>>>>>>>>    conf/bulkfiles/fbbulk-r3h.xml
>>>>>>>>>>    conf/bulkfiles/fbbulk-r4.xml
>>>>>>>>>>    conf/bulkfiles/fbbulk-r41.xml
>>>>>>>>>>    conf/bulkfiles/fbbulk-r411.xml
>>>>>>>>>>    conf/bulkfiles/fbreleases.xml
>>>>>>>>>>    conf/bulkfiles/featuresets.xml
>>>>>>>>>>    conf/bulkfiles/filesets.xml
>>>>>>>>>>    conf/bulkfiles/gbrowseconf.xml
>>>>>>>>>>    conf/bulkfiles/gbrowseconf_fb.xml
>>>>>>>>>>    conf/bulkfiles/genbanksubmit.xml
>>>>>>>>>>    conf/bulkfiles/genomeweb.xml
>>>>>>>>>>    conf/bulkfiles/organisms.xml
>>>>>>>>>>    conf/bulkfiles/sgdbulk.xml
>>>>>>>>>>    conf/bulkfiles/sgdbulk1.xml
>>>>>>>>>>    conf/bulkfiles/sgdfeatconf.xml
>>>>>>>>>>    conf/bulkfiles/site_defaults.xml
>>>>>>>>>>    conf/bulkfiles/site_eugenes_defaults.xml
>>>>>>>>>>    conf/bulkfiles/spbase.xml
>>>>>>>>>>    conf/bulkfiles/spbasefeatconf.xml
>>>>>>>>>>    conf/bulkfiles/tablewriter.xml
>>>>>>>>>>    conf/bulkfiles/toacode.xml
>>>>>>>>>>    conf/bulkfiles/tognomap.xml
>>>>>>>>>>    conf/chado2apollo-apache.conf
>>>>>>>>>>    doc/about-gff2biomart.pod
>>>>>>>>>>    doc/COPYRIGHT
>>>>>>>>>>    doc/gff2biomart-update.note
>>>>>>>>>>    doc/gmod-tools-readme.pod
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/AcodeWriter.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/BlastWriter.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/BulkWriter.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/FastaWriter.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/FeatureWriter.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/GenbankSubmitWriter.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/GnomapWriter.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/MyLargePrimarySeq.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/MySplitLocation.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/SWISS_CRC64.pm
>>>>>>>>>>    lib/Bio/GMOD/Bulkfiles/TableWriter.pm
>>>>>>>>>>    lib/Bio/GMOD/Config2.pm
>>>>>>>>>>    lib/Bio/GMOD/SeqUtils.pm
>>>>>>>>>> Please inform the author.
>>>>>>>>>> ===========================================
>>>>>>>>>>
>>>>>>>>>> When I "make", it finishes quite early, with:
>>>>>>>>>>
>>>>>>>>>> make: *** No rule to make target `bin/gmod_gff2biomart5.pl',
>>>>>>>>>> needed
>>>>>>>>>> by
>>>>>>>>>> `blib/script/gmod_gff2biomart5.pl'.  Stop.
>>>>>>>>>>
>>>>>>>>>> (the full output of "make" is below)
>>>>>>>>>>
>>>>>>>>>> Has anyone seen this problem? Is it a problem? (it seems to me...)
>>>>>>>>>> And
>>>>>>>>>> any idea what to do, please?
>>>>>>>>>>
>>>>>>>>>> Thanks
>>>>>>>>>> J
>>>>>>>>>>
>>>>>>>>>> ===========================================
>>>>>>>>>> ~/prog/chado$ make
>>>>>>>>>> Skip blib/lib/Bio/GMOD/Load.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/FeatureIO/chado.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/GMOD/DB/Config.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/Chado/Config.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/Chado/LoadDBI.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/GMOD/Load/GFF.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/GMOD/Config.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/FeatureIO/chadobulk.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/GMOD/DB/Adapter.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/GMOD/DB/Adapter/Wormbase.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/Chado/AutoDBI.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/GMOD/DB/Adapter/FeatureIterator.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/GMOD/DB/Tools/ETA.pm (unchanged)
>>>>>>>>>> Skip blib/lib/Bio/Chado/Builder.pm (unchanged)
>>>>>>>>>> make[1]: Entering directory `/home/jmalves/prog/chado/chaos-xml'
>>>>>>>>>> cp lib/Bio/Chaos/ChaosGraph.pm ../blib/lib/Bio/Chaos/ChaosGraph.pm
>>>>>>>>>> cp lib/Bio/Chaos/FeatureUtil.pm
>>>>>>>>>> ../blib/lib/Bio/Chaos/FeatureUtil.pm
>>>>>>>>>> cp lib/Bio/Chaos/XSLTHelper.pm ../blib/lib/Bio/Chaos/XSLTHelper.pm
>>>>>>>>>> cp lib/Bio/Chaos/Root.pm ../blib/lib/Bio/Chaos/Root.pm
>>>>>>>>>> cp bin/cx-download-enscore.pl
>>>>>>>>>> ../blib/script/cx-download-enscore.pl
>>>>>>>>>> /usr/bin/perl -MExtUtils::MY -e 'MY->fixin(shift)' --
>>>>>>>>>> ../blib/script/cx-download-enscore.pl
>>>>>>>>>> cp bin/cx-genbank2chaos.pl ../blib/script/cx-genbank2chaos.pl
>>>>>>>>>> /usr/bin/perl -MExtUtils::MY -e 'MY->fixin(shift)' --
>>>>>>>>>> ../blib/script/cx-genbank2chaos.pl
>>>>>>>>>> cp bin/cx-enscore2chaos.pl ../blib/script/cx-enscore2chaos.pl
>>>>>>>>>> /usr/bin/perl -MExtUtils::MY -e 'MY->fixin(shift)' --
>>>>>>>>>> ../blib/script/cx-enscore2chaos.pl
>>>>>>>>>> cp bin/cx-chadoxml2chaos.pl ../blib/script/cx-chadoxml2chaos.pl
>>>>>>>>>> /usr/bin/perl -MExtUtils::MY -e 'MY->fixin(shift)' --
>>>>>>>>>> ../blib/script/cx-chadoxml2chaos.pl
>>>>>>>>>> cp bin/cx-chaos-report.pl ../blib/script/cx-chaos-report.pl
>>>>>>>>>> /usr/bin/perl -MExtUtils::MY -e 'MY->fixin(shift)' --
>>>>>>>>>> ../blib/script/cx-chaos-report.pl
>>>>>>>>>> Manifying ../blib/man1/cx-genbank2chaos.pl.1p
>>>>>>>>>> Manifying ../blib/man1/cx-chadoxml2chaos.pl.1p
>>>>>>>>>> Manifying ../blib/man1/cx-chaos-report.pl.1p
>>>>>>>>>> Manifying ../blib/man3/Bio::Chaos::ChaosGraph.3pm
>>>>>>>>>> Manifying ../blib/man3/Bio::Chaos::FeatureUtil.3pm
>>>>>>>>>> Manifying ../blib/man3/Bio::Chaos::XSLTHelper.3pm
>>>>>>>>>> Manifying ../blib/man3/Bio::Chaos::Root.3pm
>>>>>>>>>> make[1]: Leaving directory `/home/jmalves/prog/chado/chaos-xml'
>>>>>>>>>> make: *** No rule to make target `bin/gmod_gff2biomart5.pl',
>>>>>>>>>> needed
>>>>>>>>>> by
>>>>>>>>>> `blib/script/gmod_gff2biomart5.pl'.  Stop.
>>>>>>>>>> ===========================================
>>>>>>>>>>
>>>>>>>>>> Scott Cain wrote:
>>>>>>>>>>>
>>>>>>>>>>> Hello J,
>>>>>>>>>>>
>>>>>>>>>>> I'm cc'ing this response to the schema mailing list, which is
>>>>>>>>>>> where
>>>>>>>>>>> Chado discussion usually occurs.  For the problem you are
>>>>>>>>>>> describing,
>>>>>>>>>>> there or the Apollo list (or both) would be a good place to start
>>>>>>>>>>> when
>>>>>>>>>>> looking for solutions.
>>>>>>>>>>>
>>>>>>>>>>> For using Apollo with Chado, I would suggest that you first take
>>>>>>>>>>> a
>>>>>>>>>>> look at the the tutorial on the GMOD website:
>>>>>>>>>>>
>>>>>>>>>>>  http://gmod.org/wiki/Apollo_Tutorial
>>>>>>>>>>>
>>>>>>>>>>> Additionally, I would suggest you consider using chado checked
>>>>>>>>>>> out
>>>>>>>>>>> from svn instead of the 1.0 release; it is getting pretty old at
>>>>>>>>>>> this
>>>>>>>>>>> point.  I am working on a 1.1 release that I plan to have out by
>>>>>>>>>>> the
>>>>>>>>>>> end of the year, but what is in svn now is quite stable.
>>>>>>>>>>>
>>>>>>>>>>> Scott
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On Wed, Nov 25, 2009 at 1:23 PM, J.M.P. Alves <jmalves at vcu.edu>
>>>>>>>>>>> wrote:
>>>>>>>>>>>>
>>>>>>>>>>>> Hi,
>>>>>>>>>>>>
>>>>>>>>>>>> I am following the instructions from README.Apollo in the GMOD
>>>>>>>>>>>> 1.0
>>>>>>>>>>>> distribution, and have found a couple of problems.
>>>>>>>>>>>>
>>>>>>>>>>>> When running
>>>>>>>>>>>>
>>>>>>>>>>>> psql chadodb < modules/sequence/apollo-bridge/apollo.inserts
>>>>>>>>>>>>
>>>>>>>>>>>> I got a couple of errors. The first:
>>>>>>>>>>>>
>>>>>>>>>>>> ERROR: null value in column "cv_id" violates not-null constraint
>>>>>>>>>>>>
>>>>>>>>>>>> I checked the offending part of the sql and it tries to find a
>>>>>>>>>>>> cv_if
>>>>>>>>>>>> for the
>>>>>>>>>>>> record with "Relationship Ontology" as name. But in the schema I
>>>>>>>>>>>> installed
>>>>>>>>>>>> there is no such record. The closest I could find was one called
>>>>>>>>>>>> "relationship" and one called "pub relationship type". Which is
>>>>>>>>>>>> the
>>>>>>>>>>>> correct
>>>>>>>>>>>> one?
>>>>>>>>>>>>
>>>>>>>>>>>> The second error was:
>>>>>>>>>>>>
>>>>>>>>>>>> ERROR: aggregates not allowed in WHERE clause
>>>>>>>>>>>>
>>>>>>>>>>>> This was triggered by the command:
>>>>>>>>>>>>
>>>>>>>>>>>> delete from  feature_namegenerator where
>>>>>>>>>>>> feature_namegenerator_id
>>>>>>>>>>>> not
>>>>>>>>>>>> in
>>>>>>>>>>>> (select feature_namegenerator_id from feature_namegenerator
>>>>>>>>>>>> where
>>>>>>>>>>>> count
>>>>>>>>>>>> =
>>>>>>>>>>>> max(count) group by name,type_id);
>>>>>>>>>>>>
>>>>>>>>>>>> I haven't figured out how to solve this one, since I have no
>>>>>>>>>>>> experience
>>>>>>>>>>>> with
>>>>>>>>>>>> PostgreSQL.
>>>>>>>>>>>>
>>>>>>>>>>>> Regards
>>>>>>>>>>>> J
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> P.S.: I wrote a few minutes ago about a problem with GBrowse's
>>>>>>>>>>>> Chado
>>>>>>>>>>>> documentation. I spoke too soon, there was another problem that
>>>>>>>>>>>> I
>>>>>>>>>>>> found
>>>>>>>>>>>> when
>>>>>>>>>>>> I clicked on the examples in GBrowse (gene name). Adding the
>>>>>>>>>>>> following
>>>>>>>>>>>> table
>>>>>>>>>>>> to the permissions solved it:
>>>>>>>>>>>>
>>>>>>>>>>>> GRANT SELECT ON all_feature_names TO "www-data";
>>>>>>>>>>>>
>>>>>>>>>>>> --
>>>>>>>>>>>> -------------------------------
>>>>>>>>>>>> João Marcelo Pereira Alves (J)
>>>>>>>>>>>> Post-doctoral fellow
>>>>>>>>>>>> MCV / VCU - Richmond, VA
>>>>>>>>>>>> http://bioinfo.lpb.mic.vcu.edu
>>>>>>>>>>>> f. 1-804-828-3897
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> -------------------------------
>>>>>>>>>> João Marcelo Pereira Alves (J)
>>>>>>>>>> Post-doctoral fellow
>>>>>>>>>> MCV / VCU - Richmond, VA
>>>>>>>>>> http://bioinfo.lpb.mic.vcu.edu
>>>>>>>>>> f. 1-804-828-3897
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> ------------------------------------------------------------------------------
>>>>>>>>>> Join us December 9, 2009 for the Red Hat Virtual Experience,
>>>>>>>>>> a free event focused on virtualization and cloud computing.
>>>>>>>>>> Attend in-depth sessions from your desk. Your couch. Anywhere.
>>>>>>>>>> http://p.sf.net/sfu/redhat-sfdev2dev
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Gmod-schema mailing list
>>>>>>>>>> Gmod-schema at lists.sourceforge.net
>>>>>>>>>> https://lists.sourceforge.net/lists/listinfo/gmod-schema
>>>>>>>>>>
>>>>>>>> --
>>>>>>>> -------------------------------
>>>>>>>> João Marcelo Pereira Alves (J)
>>>>>>>> Post-doctoral fellow
>>>>>>>> MCV / VCU - Richmond, VA
>>>>>>>> http://bioinfo.lpb.mic.vcu.edu
>>>>>>>> f. 1-804-828-3897
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>> --
>>>>>> -------------------------------
>>>>>> João Marcelo Pereira Alves (J)
>>>>>> Post-doctoral fellow
>>>>>> MCV / VCU - Richmond, VA
>>>>>> http://bioinfo.lpb.mic.vcu.edu
>>>>>> f. 1-804-828-3897
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>> --
>>>> -------------------------------
>>>> João Marcelo Pereira Alves (J)
>>>> Post-doctoral fellow
>>>> MCV / VCU - Richmond, VA
>>>> http://bioinfo.lpb.mic.vcu.edu
>>>> f. 1-804-828-3897
>>>>
>>>>
>>>
>>>
>>>
>>
>
> --
> -------------------------------
> João Marcelo Pereira Alves (J)
> Post-doctoral fellow
> MCV / VCU - Richmond, VA
> http://bioinfo.lpb.mic.vcu.edu
> f. 1-804-828-3897
>
>



-- 
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

More information about the Gmod-help mailing list