Readseq was only designed to handle personal sequence files,
not database sized files. I DO NOT recommend its use for
database distribution files from PIR, GenBank, EMBL, etc.
Having said that, if you do split the PIR files into, say, 10mb
chunks (a few hundred sequences each perhaps), readseq may work
okay. It should understand and translate from PIR/CODATA to
NBRF. However, you will loose any documentation from the
PIR sequences. Readseq still just drops any documentation
sections it comes across, because teaching it to translate
from one documentation format to another is a large job.
-- d.gilbert--biocomputing--indiana u--bloomington--gilbertd at bio.indiana.edu