[Bioperl-l] Failed to load tigr xml files through Bio::Seq::tigr

[Jason Stajich]

Josh will have to chime in....

I don't know that Bio::SeqIO::tigr can parse all the different  
flavors from TIGR, but I believe it was written particularly for the  
rice data and maybe not for the version 3.  Can you figure out what  
is different in the version 3 over previous version of the TIGR  
format? If I remember correctly the module was written with regexps  
instead of an XML parser so I think it is pretty brittle to changes.   
We may want to  try and re-write it with XML::SAX or another XML  
parser - I have written tigrxml for a different TIGR XML flavor in  
this way.  But we need someone to volunteer to help do this sort of  
thing....

In this case I think you need to submit a bug rather than posting the  
huge file to the mailing list. http://bugzilla.open-bio.org

-jason


On Nov 2, 2005, at 5:36 PM, Kun He wrote:

> Dear all,
>
> I have just tried to use the Bio::Seq::tigr module (BioPerl 1.5.1)  
> to import information from TIGR Rice version 3 XML format  
> pseudochromosome data, but it always failed on the loading  
> (initiating) procedure.
>
> "
> use strict;
> use Bio::SeqIO;
>
> my $file=shift;
>
> print $file,"\n";
> my $io = Bio::SeqIO->new(-file => "<$file", -format => "tigr" );
> "
>
> There are a lot kinds of errors thrown out, based on different XML  
> files. Attached is a sample xml from tigr, who can kindly help me  
> to try if you  can load it by the tigr.pm module? Thanks a lot!
>
> Kun He
>
[snip out XML, save bits!]

--
Jason Stajich
Duke University
http://www.duke.edu/~jes12