I've got Wu-Blast running on a Linux box (Redhat 7 Kernel 2.2.16-22enterprise on
a 4-processor i686) and I seem to be able
to take a nucleotide sequence and run blastx against genbank's nr database just
fine.
But when I try to run a tblastx against the nt database (also downloaded from
Genbank) I get the following error message:
WARNING: The associated FASTA-format database file is too large and will be
ignored: /usr/ncbi/blast/db/nt
Strangely enough, though, it still gives me results in my output file.
Not knowing if I should worry about this error message, I ran setdb on nt to try
to create a protein database from nt,
and then ran blastx against it with my nucelotide sequence - and I got no
matches (where I did get matches from the tblastx, even though it tells me its
ignoring the database.)
I guess I'm wondering two things:
1. Should I worry about this error message? (In other words, can I trust the
results its giving me, despite this error message)
2. If I have to worry about it (which for now I assume I do) is there any way to
get around this?
I didn't set up the Linux machine myself, and I'm not sure how I can
double-check my swap space, but when the machine was set up
they did ask me how much swap space I wanted, and I asked for 2 Gig. I'm
assuming that's what's there.
If anyone can help me with this, I'd greatly appreciate it!
Thanks!
Betastar
---
