[BioSQL-l] Importing GFF3 files into a BioSQL database?

[James Procter]

Hi Peter.

Peter wrote:
>> I'd like to stick my oar in here and say that regularising the
>> GFF3<>BioSQL mapping accross Bio* is an essential step towards
>> round-tripping between the different language bindings, ...
> 
> Absolutely - that's why I started this thread.  On a related note,
> there is probably still room for improvement in the GenBank<>BioSQL
> mapping across the Bio* projects, where the "gold standard" also needs
> documenting.
definitely. An awful lot of effort has been expended on standardisation
in the last 4-5 years, so it shouldn't be a tough job for someone with
experience of the GenBank annotation model.
<SNIP>
>> Its very rough and ready, probably biased and not quite correct at the
>> moment. It does, however, provide a central home for the information
>> that Brad has put up on his blog, and the various other notes that have
>> been made in this thread. Please add and amend as appropriate.
> 
> Is this based on the current behaviour of BioPerl code - or just what
> you think would be sensible?
Current behaviour, where possible, with notes where current behaviour
seems broken/crazy, based on my mishmash of experience over the last
couple of months with nightly builds of the 1.6 code. You'll also notice
that I've tried to stick to the well defined basics, leaving the detail
to be filled in by the experts.

The aim of the page is to allow some consensus regarding each aspect of
mapping the GFF spec to be reached, and thence enable us to verify or
note where each language binding differs from that consensus.

I'm interested in this because I'm exploring the DAS<Bio*>BioSQL mapping
in a situation where I need to import via BioPerl, and retrieve features
via BioJava. I'll add in my own notes for this special case (based on
biojavax bindings and an extensively patched version of Dazzle's BioSQL
datasource) over the next few weeks. I'll also try to encourage some
input from other DAS developers at the next DAS Developers meeting in March.

Jim.