[Bioperl-l] Stop him before he codes again! [almost-multiple-alignment tool?]

[George Hartzell]

Yep, there's nothing scarier than some people at the keyboard....

I have:

  1     putative sequence (DNA, not too big 1-5kb).
  many  (1-10) empirically determined sequences (DNA) that should be
        fairly similar to the putative sequence.

It's trivial to:

  produce a pairwise alignment of each empirical sequence against the
  putative sequence.

I'd ultimately like to produce (actually, I don't have much choice...):

  an almost multiple-alignment-like figure, with the putative sequence
  (e.g.) along the bottom and the empirical sequences piled up above
  it, gapped in emp. and put. sequences where necessary.

It's pretty much similar to piling up a bunch of EST's/cDNA vs. the
corresponding genomic, with a simpler gapping model (no splicing,
etc...).

Ideally I'd like to just wave my hands and have it all work :).  I'll
write code if I have to....

Since I'm not really looking for a multiple alignment, I'd like to
avoid the cost of actually computing/approximating one.

I'd like to just do the pairwise alignments then shoehorn the results
into an existing bioperl multiple-alignment representation and play
with it from there.

I'd love comments (and will happily pay in beer/coffee next time
you're in Berkeley) about:

    - existing tools that do just this (or even close)

    - what's the cleanest bioperl object to shoehorn it into?
      SimpleAlign?  Align?

    - given one of those objects, are where should I start digging for
      pretty output routines?

    - if there's nothing particularly useful, any suggestions on how
      to structure things so that I can deposit them into BioPerl?

    - should i just punt and run something (e.g. clustalw, pileup)
      against the putative and all the empirical and be done with it?
      Tool recommendations?

Thanks,

g.