[Bioperl-l] Genebank mRNAs

[Ewan Birney]

On Fri, 4 May 2001, Jason Stajich wrote:

> John -
> 
> Can you please submit example code and genbank accession numbers that cause
> the behavior you describe in your bug reports.  It will greatly increase the
> chances that someone will make the time to fix these bugs if we don't have
> to spend time tracking down an example.  Ideally  we'd want to design a test
> case for the bug in the t/SeqIO.t test file so we can properly test for it.
> If you wanted to do this I know it will take one of us a lot less time to
> fix it.

John - just to reiterate - if you give us a file which we screw up on we
will fix the bug. It is almost impossible to debug things without a
reproducible bug... ;)



[warning to people who aren't interested in the Guts of bioperl and just
want to have things "work" - this is one of those boring object
discussions]


John's other "bug", that SeqFeature::Generic->seq should return the
spliced DNA is a deeper problem. This is not a current "bug" - just that
we don't do this. 


My view here - for what it is worth - is that the ideal solution is to
automatically "promote" mRNA and CDS lines into a GeneI/TranscriptI object
on which one can ask for the mrna (or whatever the method call
is) sequence.


I think handling this at the SeqFeature/SeqLocation level is more
dangerous - how do we seriously honour the fuzzy location stuff down there
in the midst of returing the "spliced" DNA (I guess we can't). 



Anyway - John's problem - which is that you can't easily get out the
spliced DNA should be addressed by us for the next round of
improvements... 






-----------------------------------------------------------------
Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
<birney@ebi.ac.uk>. 
-----------------------------------------------------------------