I use Blast 1.4.8 and the threaded Wash-U 1.4.9MP Blast executables on
an Alpha 2100/A system running Digital Unix 3.2D
Both stopped working (or rather started producing strange output)
roughly about the time of the new Genbank release.
Both versions of BlastN that I have work on an older genbank with
967,554 total sequences, but don't work when searching a newer file
containing 970,663 total sequences. I've rebuilt the binary indices
and the result has been the same.
Fasta 3.0t has no problem searching the same datafile.
Has anyone else had this problem with Blast? Unix GCG Blastn had
a similar problem and our GCG admin had to break off the EST
sequences into a seperate DB. I would prefer not to have to do the
cdagdigian at genetics.com