[EMBOSS] case sensitive identifiers - Checked by AntiVir DEMO version -

[pmr at ebi.ac.uk]

Dear Guy,

> So, your idea of fetching all entries and then parsing them would work
> for SRS. I however think that instead of an associated parameter
> -scasesensitive it would be better to have in the emboss.default syntax
> for DB entries an optional parameter case:. You should be able to handle
> the situation where it is appropriate to pass an id to a case sensitive
> search method and the situation where it is appropriate to parse the
> output of a case-insensitive search method. This can best be decided for
> each
> databank at EMBOSS site configuartion time, rather than at sequence
> retrieval time. What do you think ?

You are right. Excellent idea. I will do it ASAP.

It will mean a new attribute stored for sequence database access.

For the PDB case, really only the end of the ID is case-sensitive. Do you
think the database should be case-sensitive for the whole ID, or does it
make sense to check for a pattern as the case-sensitive part?

EMBOSS will initially read only one sequence for a seqall ... it does not
read in all the sequences and look for duplicates so we have to decide in
the emboss.defaults DB definition how to check a single ID (no way to read
them all and check for duplicates).

regardrs,

Peter