[BiO BB] A series of Pairwise alignments

Mike Marchywka mmarchywka at eyewonder.com
Sat Jun 3 10:23:53 EDT 2006


Yeah, it occured to me the objective would drive the definitions.
I've been thinking about stuff like this from the immune's system point of view.
Consider something like clustering software since with that many
sequences they probably can be lumped into clusters. It occured to me you could
take each sequence, break it up (generate "words" (peptides) using your favorite epitope
 prediction code ), and cluster the sequences based on "vocabulary".
This may tell you which groups of proteins look similar to the immune system.

btw, has anyone used tools or features like the "signatures" mentioned here?
http://www.bioinf.man.ac.uk/dbbrowser/PRINTS/printscontents.html#Receptors

*************************************************************************


-----Original Message-----
From:
bio_bulletin_board-bounces+mmarchywka=eyewonder.com at bioinformatics.org
[mailto:bio_bulletin_board-bounces+mmarchywka=eyewonder.com at bioinformati
cs.org]On Behalf Of Ann Loraine
Sent: SaturdayJune-03-2006 10:00 AM
To: The general forum at Bioinformatics.Org
Subject: Re: [BiO BB] A series of Pairwise alignments


Howdy,

Probably you would want to run a pairwise alignment program over each
pair of seqs and then write some wrapper code around that to process
the data as you like.

The alignment program you select should be something that fits your
goal - do you want global or local alignment? What kind of scoring
system do you need? And so on.

For fun, take a look at:

http://helix.biology.mcmaster.ca/721/outline2/node42.html

-Ann
\



More information about the BBB mailing list