From: Dr. med. Christoph Gille <christoph.gille@ch...> - 2014-01-23 13:31:43
Currently I am using Blat and would like to know whether Bfast would be
better. Bfast keeps the database indices in a file, whereas Blat calculates
indices each time.
Therefore I assume that Bfast would run faster.
My question: Is Bfast better suited for the following task?
I have a bunch of peptide sequences given as one-letter-amino acid text and
want to find the respective
entry in a local sequence DB (1,000,000 sequences as fasta file).
I could use /usr/bin/fgrep which is of course not efficient.
Currently, I reduce the number of letters (20 different AA) to 4 (ACTG) prior
searching with Blat.
I consider only the forward match as the rev complement strand does not make
Blat is very good at it. It takes only a few sec.
Would Bfast be faster as its name suggests?
Btw, even though Bfast compiled on my Debian 7, the tests failed. I had
to compile on Debian 6 and move the executable to the Debian 7
machine. This executable worked.
This was the case for versions 0.7.0 and 0.6.5
Many thanks for your help