[Biodevelopers] Re: BLAST asymmetrical

[Titus Brown]

On Tue, Jan 23, 2007 at 07:25:37AM -0800, Martin Gollery wrote:
-> My apologies, Chris, I see what you are saying and you are correct-
-> S/W is not a good program for genome/genome comparisons because it
-> does not show all the sub-optimal alignments. Somebody modified it so
-> that it would do this, but I can't recall the name.
-> 
-> Double-affine S/W will do much better at this sort of thing, but it
-> fails to catch rearrangements- this is not what it is designed for.
-> Double-affine S/W would be ok if the genomes are close, but not
-> otherwise.

There is a large literature on the subject of aligning genomes that is
not being mentioned here ;).  For animal genomes, I would suggest
looking at the blastz and MultiPipMaker work from Penn State in
particular, as well as the VISTA work at LBL.  I'm less familiar with
whole-genome microbial analyses but then again they're relatively
tractable (computationally)...

E-mail me if you need specific refs.

cheers,
--titus