[Biojava-l] BioSQL

[David Huen]

On Thursday 20 Jan 2005 13:46, David Lapointe wrote:
> In BioJavaInAnger David Huen has the disclaimer at the end of his useful
> section
>
> NOTE: If you are using the 1.3 version of Biojava with the Singapore
> schema, do not install biosqldb-assembly-pg.sql or
> biosql-accelerators-pg.sql as described above. All you will need is the
> the new biosqldb-pg.sql. There appear to be performance issues in some
> cases when the other stuff is installed also. This note will be updated
> eventually to reflect this advice.
>
> It is not clear to me which way to go.  How does one know whether one has
> the SIngapore schema ?

I believe that the current schemas are at:-
http://cvs.open-bio.org/cgi-bin/viewcvs/viewcvs.cgi/biosql-schema/sql/?cvsroot=biosql

The last time I tried with postgres,  I believe the biosqldb-pg.sql alone 
was sufficient (i am relying on hazy memories here, anyone else know 
better?) and the drop-tables.sql is useful for cleaning up.  Performance 
was quite slow but Thomas Down found that introducing a further index 
improved things:-

create index seqfeatureloc_seqfeature on location (seqfeature_id);

I don't know if this has been committed since.  If you have performance 
problems with postgres, do investigate adding indices to the postgres case.

Regards,
David H.