Blast cut-offs

[Keith Bradnam]

This is sort of loosely related to acedb, so I hope people don't mind me
raising it here.

When people integrate DNA and protein homologies into their acedb
databases, do people use any cut-offs when generating the BLAST results?
I.e. do people restrict BLAST searches to just those above a certain score
or E value?

I'm thinking about whether to cull some of the homologies in our database
which have low BLAST scores, but there's always the worry that even weak
homology information might be of use to someone.

I guess that this is sort of related to the more difficult question of
'what is a useful BLAST result?' which would largely depend on what the
experimenter considers useful.

Anyway, I would welcome comments from anybody who manages a lot of BLAST
information in their databases.

Thanks in advance,

Keith

~  Keith Bradnam - Developer, Arabidopsis Genome Resource (AGR)
~  Nottingham Arabidopsis Stock Centre - http://nasc.nott.ac.uk/
~  University Park, University of Nottingham, NG7 2RD, UK
~  Tel: (0115) 951 3091