hash function for 64 codons

Tim Cutts timc at chiark.greenend.org.uk
Mon Jun 18 03:55:29 EST 2001


In article <etdvglzf04l.fsf at lola-granola.mit.edu>,
Alex  <new_name at mit.edu> wrote:
>
>> I guess there is probably a good hash function out there that hashes
>> the 64 codons, which I assume is the fastest way to do translation.
>
>I think a 4x4x4 lookup table is faster, actually.

I agree.  I have some C code which uses this method (in fact, two 4x4x4
tables) to translate all six reading frames in a single pass over the
input DNA sequence (no reverse complementation step is therefore
necessary).  I think it's pretty much as fast as I can make it.
It's part of my tpatterns program, which you can download from my web
site:

http://www.cyclin.demon.co.uk/

Hope you find it useful.

Tim.

---





More information about the Bio-soft mailing list