Having trouble converting genbank format to gff format

[jmartin]

I have a fairly large number of sequin annotation files generated by GenBank's automated PGAP pipeline that I want to convert into bed format. I've been able to use the asn2gb program from the ncbi toolkit to generate GenBank format files from those sequin files, but so far I haven't had much luck converting from genbank into gff. After getting into gff, conversion into bed won't be a problem. But so far I haven't had any luck going from genbank to gff.

I am trying to use bp_genbank2gff3.pl, but am getting this error message:

--------------------- WARNING ---------------------
MSG: Bad LOCUS name? Changing [Staphylococcus_spHMPREF3292-1.0_Cont0>11789] to 'unknown' and length to Staphylococcus_spHMPREF3292-1.0_Cont0>11789
---------------------------------------------------

I thought maybe it was unhappy because of the greater-than symbol in the LOCUS line. So I tried removing it, and then I get a new error message:

------------- EXCEPTION: Bio::Root::Exception -------------
MSG: asking for tag value that does not exist date
STACK: Error::throw
STACK: Bio::Root::Root::throw /gsc/scripts/opt/genome/current/user/lib/perl/Bio/Root/Root.pm:357
STACK: Bio::SeqFeature::Generic::get_tag_values /gsc/scripts/opt/genome/current/user/lib/perl/Bio/SeqFeature/Generic.pm:498
STACK: main::gff_header /gsc/bin/bp_genbank2gff3.pl:895
STACK: /gsc/bin/bp_genbank2gff3.pl:406
-----------------------------------------------------------


All my input files were generated by GenBank itself, and I used a GenBank script to generate the .gbk files I now have. So I would think my .gbk files are correctly formatted. Can anyone either advise me on what I'm doing wrong with the 'bp_genbank2gff3.pl' script? Or if there is a better way to convert from .gbk into .gff that would work too.


Thanks,
John Martin

[pmiguel]

The "bp_" prefix stands for "BioPerl". You might want to ask the BioPerl programmers what the problem might be. They have an email list. Possibly there are other ways to query them as well.

--
Phillip

[colindaven]

From memory I think readseq.jar might be able to help, it is simple too.

http://iubio.bio.indiana.edu/soft/molbio/readseq/java/