[Bioperl-l] Bioxml Parser.

Bradley Marshall bradmars@yahoo.com
Sun, 26 Nov 2000 13:01:35 -0800 (PST)

>Could you be more specific about what you think
should rather go to
>their own methods? Sorry for not looking at the code
directly :-)

I assume you mean in extending the SeqFeature object?

The way that game organizes it features is either by
the computation performed (say a single blast run) or
by the annotation group (so maybe all the exons of a
gene).  Much of the details are stored at the
computation or annotation level rather than at the
analysis level.  So when parsing into bioperl I have
to store all the data about the computation performed
in each SeqFeature.  I was thinking about making some
type of feature collection that could have it's own
methods containing the details about each computation.
 I have to look more closely at how, say, the
GeneCollection is implemented.

The other thing that game has is the idea of a
seq_relationship.  So you can specify an alignment
between two sequences.  I think this is similar to
bioperl's FeaturePair.  So maybe I should shoot for a
collection of FeaturePairs?

>Second, out of curiosity (please don't flame me for
my ignorance):
>what's the specific difference to SeqIO::GAMEHandler?
I'm not even sure
>the latter is actively maintained ...

The old SeqIO::bioxml and SeqIO::GAMEHandler are
outdated.  I'll remove them.


Do You Yahoo!?
Yahoo! Shopping - Thousands of Stores. Millions of Products.