aligning protein alternative splicing isoforms

[Q S Doctor]

What parameters would you suggest for aligning protein isoforms that
(presumably) come from alternative splicing.

In this case, I don't have the mRNA or genes to map back to. I will
use the assumption that regions less than ~100% identity should not be
aligned; however we do expect some SNP substitutions.

Is T-Coffee up to this? Or better to use some thing specifically
engineered for that application? Are there other tools geared for this?

[tcoffee msa]

It is notoriously difficult to align alternative splice forms and I
do
not know of any tool doing very well
1-adding local information:

t_coffee <seq> -method lalign_od_pair proba_pair

2-combining methods better designed to deal with local similarity:
t_coffee <seq> -method poa_pair, dialignt_pair, (fsa_pair)

I am not sure about fsa, as I have never validated it. All these
methods
may fail in the case of repeated domains contained in mutually
exclusive
exons (i.e. the program will try to align regions of homology).

Cheers,

Cedric