[Bioperl-l] Aligning two alignments ...
bix at sendu.me.uk
Tue Mar 27 13:24:03 EDT 2007
Stéphane Téletchéa wrote:
> Dear all,
> I'm trying to compare alignments results for small protein sequences to
> validate my result from another method. In the best option, i would do
> on two sequences
> - run a clustalw alignment
> - run a blast alignment
> - run a T-Coffee alignment
> - use my own
> And align them all to see easily what is identical, what differs.
You don't want to align the alignments; that makes no sense and you
can't achieve anything without altering the alignments.
I suppose you actually just want to score the alignments against some
'ideal' alignment or against one of the others. In which case generate a
match line/ cigar line/ simply count the number of identical bases
across different alignment strings.
More information about the Bioperl-l