[Bioperl-l] Displaying sequence with Tk: Advice?

[David Block]

Hello all!

I'm looking for some advice/suggestions from possible end-users.  We're
deep in the middle of coding (Mark is trying to decipher BioTk's zooming
routine), and we need some input:

What would be the most helpful way to display sequence
features?

Given: an axis in the middle of the canvas, with numbers
referring to the # of base pairs relative to the start of the sequence.  
Given: sequence features above/below (left/right if vertical) referring to
forward/reverse strand features.
Given: different offsets from the axis, and different colours for
different types of features (could be repeats, UTRs, etc., or exons from
different sources (Genscan, Grail, Blast homology)).

Problem: Multiple levels of features/sub-features.  Do we make genes just
another type of feature, with exons in another row?  Or do we mark genes a
la Acedb with connector lines connecting exons? Or do we create two maps,
the one just for top level features, and the second for lower level
features like exons, etc.?

Problem: Annotation vs just display.  We envision a blank canvas for the
annotator to transfer features onto, creating genes out of the parts
provided from databases, prediction algorithms, homologies, etc., and then
editing those pieces to remove stop codons, etc.  Do we just add rows to
the canvas for 'my accepted genes/features', or do we create an additional
map for annotations?
<warning>two maps for reading/annotation * two maps for toplevel/lower
level = 4 maps on your screen.  Too many? </warning>

Weigh in, please!

Dave "Now I'm waiting for Mark to finish the zoom scaling" Block