[Bioperl-l] Another update for my alignment module

Yee Man Chan ymc at paxil.stanford.edu
Mon May 19 17:29:43 EDT 2003


Hi Ewan,

	I got another update for my align module. This time I stole some
ideas from the align routine in ssearch to make my code faster. This makes 
it 9.35% faster than the align routine for ssearch because I am using
gap arrays to represent alignment and I used pointers to walk the Gotoh
arrays. You can find my new code at:

http://www.stanford.edu/~yeeman/dsw.tgz

	I believe my code generates correct alignments because I ran it
through some 1700 pairs of 100kb sequences and it seems to show very
similar result to AVID.

	The search phase used in my algorithm is still Miller-Myers (The
fastest Smith-Waterman that calculates every cell). The search phase time
can be reduced in half if I replace it by Phil Green's algorithm (an even 
faster one but it ignores some cells based on some heuristics.). As a
future work, I can add Phil Green's code to my module for people who want
to sacrifice some sensitivity for speed.

	I think my code is now worthy to be included in bioperl. Can you
tell me how I can go about doing it?

Thanks a lot.
Yee Man



More information about the Bioperl-l mailing list