[Biojava-l] Question on FlexibleAlignment

[Eric Haugen]

The idea behind the UnequalLengthAlignment interface implemented by FlexibleAlignment is that you don't need to have terminal gaps in any sequence, you can just set the location's min to be the position in the alignment where the sequence's first symbol aligns.

But I haven't tried using it in that way, so I don't know how well it actually works...

-- 
Eric Haugen
Software Engineer
University of Washington Genome Center


On Wed, 11 Apr 2007, Russ Kepler wrote:

> I've re-encountered something that has always puzzled me - I'm building a
> FlexibleAlignment with a bunch of SimpleAlignmentElements composed of
> SimpleGappedSequences.  The SimpleAlignmentElement wants a RangeLocation, but
> insists that the max()-min()+1 be sequence.length().  But the
> sequence.length() includes any gap placed at the start of the sequence, so
> necessarily you include any offseting gap to make the alignment.  This means
> that FlexibleAlignment.getLabelsAt(1) will always return the list of all the
> SimpleAlignmentElements.  It seems that either the getLabelsAt() should
> exclude the elements that don't have sequence at the position (yet) or the
> SimpleAlignmentElement shouldn't be using the length of the sequence in
> testing the RangeLocation.
>
> Any comments?  I hope this makes sense to someone...
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biojava-l
>