[Bioperl-l] blastall and blosum62 problem

[Ian Korf]

>
> But remember this: when I try blastn I get good results again:
>
> cat sequences.in | blastall  -p blastn -d /usr/local/blast/data/db1 -M 
> BLOSUM62 | more
>

blastn doesn't use scoring matrices, it uses match/mismatch values (-r 
and -q). So the -M BLOSUM62 is probably being ignored. If you don't 
have a .ncbirc file, you can also set the BLASTMAT environment variable 
to the directory containing your matrices.

-Ian