[Bioperl-l] new GFF2 parsing/dumping routines committed...

Mark Wilkinson mwilkinson@gene.pbi.nrc.ca
Tue, 21 Nov 2000 08:57:58 -0600

> I am pretty sure that
>    (A) at least 75% of GFF will not be strict GFF2 compliant

yeah, probably not... this is why I didn't want to touch the existing routines, which are a bit more lenient  ;-)

Having said that, making Feature::Generic GFF2 compliant might now *encourage* people to follow the suggested standard... I agree that GFF2, as
described on the Sanger website, is a "good thing", and is really surprisingly powerful if you follow the rules!!

In any case, after giving the Feature object the ability to dump itself in this strict format (for selfish reasons), it was obvious that one
should expect to be able to re-create the identical feature object ab initio from the GFF dump of that feature... so it became necessary to
write the Generic Feature import routine to be GFF2 compliant also...

Is it true that ACeDB dumps features with "strict" GFF2 compliance??  I should drop Ed a line and ask him... If so, these changes will make
life easier all-round!!

Anyway... happy BioPerling all!


Dr. Mark Wilkinson
Bioinformatics Group
National Research Council of Canada
Plant Biotechnology Institute
110 Gymnasium Place
Saskatoon, SK