Farm files for databases

[simon andrews (BI)]

I'm trying to find the best way to do the following:

I have an application which returns an identifier (effectively an accession
number), which could be present in any one of 4 separate EMBOSS databases.
I'd like to be able to search all of these databases and retrieve the
sequence from whichever one finds it (I know that the identifiers are unique
between the different databases - so I'll only ever find one entry).

Having read the EMBOSS documentation the only reference I could find for
doing this sort of thing was to make a database entry with an "EXTERNAL"
format, and then have seqret query a script to return the sequence.  However
the details for this are pretty sketchy.

What exactly would a script of this type have to do?  What input is it
supplied with (and how), and what must it return?  Is this the only (or
best) way to do what I'm trying to do?

Any help is much appreciated.

	TTFN

	Simon.

----
Simon Andrews PhD
Bioinformatics Dept
The Babraham Institute

simon.andrews at bbsrc.ac.uk
+44 (0)1223 496463