[Bioperl-l] Blast or blat against custom db?

[Kevin Brown]

I've been tasked to write a "small" application for the lab I work at
that basically starts with an NCBI file for an organism and goes through
a number of steps to distill out the unique protein coding sequences and
then designs oligos for the building of the genes.  One of the steps is
comparing the overlap region of the oligos to all the others designed to
try and prevent mismatches in the build that might truncate a gene or
splice in another gene into it during the build step.  I tried to do
this within perl with just a looped string comparison regex, but by my
calculations comparing each half of an oligo with all the other oligos
for this organism results in well over 8 BILLION comparisons needed.
The system was still crunching at it 3 days later with no sign of
nearing completion.

So, my thought was to utilize something like blastall from within the
script to find other oligos of similar match, but it means that I need
to dump out the oligos designed, create the db with formatdb.  Then do
the blast and finally analyze the result file to see what needs to be
changed in the oligos to prevent a mismatch redesigning any matches.
I'm just trying to figure out how to do it all without leaving the
script, but as yet haven't noticed a way to create a db from within perl
using bioperl?

Any thoughts on directions I should look?