QN: Tell me about this GenBank search algorithm.

Peter Rice pmr at sanger.ac.uk
Thu Apr 27 09:21:30 EST 1995


In article <3nk69m$bpp at goanna.cs.rmit.edu.au> hugh at goanna.cs.rmit.edu.au (Hugh Evan Williams) writes:
>   A friend posted me some mail recently concerning an ABI seminar he
>   attended in which they discussed (in true marketing jargon) a program
>   which claims to be able to search GenBank in 2 minutes with a "single
>   based FDF with 480 processors" (i get the impression this is some kind
>   virtual pipeline technology thing [though i'm not even sure what that is]
>   in which a computer has virtual processors, in a concept similar (?) to
>   virtual memory).
>
>   The program has three search methods: "FDF sliding window", "Smith-Waterman"
>   and FASTA. I am well versed in the latter two, but not in the first. Does
>   anyone know anything else about this FDF stuff? Does this software really
>   do a Smith-Waterman alignment on the whole of GenBank (seems doubtful)?

Sounds like the "Fast Data Finder" chip. The search time means very little
without knowing the query length (and the age of the GenBank release. Also
remember that a database search using Smith-Waterman only needs the scores,
not the full alignment, so there is a saving there. It should be very fast.

There are other ways to do a full Smith-Waterman on a sequence database.
I have worked with implementations on MasPar (EMBL's BLITZ server) and
on the Bioccelerator. There are quite a few more around too.
--
------------------------------------------------------------------------
Peter Rice                           | Informatics Division
E-mail: pmr at sanger.ac.uk             | The Sanger Centre
Tel: (44) 1223 494967                | Hinxton Hall, Hinxton,
Fax: (44) 1223 494919                | Cambs, CB10 1RQ
URL: http://www.sanger.ac.uk/~pmr/   | England




More information about the Bio-soft mailing list