[Bioperl-l] .qual are like fasta files

Ewan Birney birney@ebi.ac.uk
Tue, 7 Aug 2001 08:19:16 +0100 (BST)


Chad -


All this sounds great! Good stuff!



i think the deliberate copying of PrimarySeq to PrimaryQual is fine. I
would put it in 

Bio::Seq::PrimaryQual

so we don't mess up Bio::  namespace (despite the fact that
Bio::PrimarySeq is there).

there is an argument that although PrimaryQual should basically look like
a copy and paste, it should in fact use

  $obj->qual() # equivalent to ->seq, returns an array of numbers

and

  $obj->subqual(10,20) # equivalent to ->subseq

as it is not really a sequence and we will mislead people less.



I suspect inherieting from Bio::PrimarySeqI is probably a bad idea. open
quesiton what one does with trunc. translate is not possible ;) 


We really need a Bio::IdentifiableI interface that it could inheriet for
the identifier set.




After than I would make a 


Bio::Seq::SeqWithQuality


which would inheriet from Bio::Seq (hence have a Bio::PrimarySeq and
all the other goodies) but have in addition a Bio::Seq::PrimaryQual, and
give access to the quality values in many places. 


(ewan spots a rather annoying memory circle there and somewhat curses his
luck. - to do with Bio::SeqFeatureI->entire_seq giving back a
Bio::PrimarySeq but ideal in this case you would want access to the 
Qual values as well but then if it pointed to the parent you would
end up with a cycle. hmmmmm)


Bio::SeqIO::phred  could then make a Bio::Seq::SeqWithQuality object





What do other people think?




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