Hi Arek and Richard,
Thank you both very much for your responses; I'm replying to both in
one email. The database is indeed centered around reagents; in fact,
reagents and locations - the database serves as a data storage for a
LIMS system I'm building for our laboratory. I've managed to create 2
datasets from tables reagents, containers, wells, experiments and
preps - resulting in exactly the concept I want. The next issue that
I haven't resolved quite yet is Martview search - I am not getting all
of the filters that I want for the above datasets. I suspect
configuration might be one of the reasons (if not the only one) -
therefore, I am currently going over the configuration steps again in
attempt to resolve this problem. If you do get a chance, however, to
look at it yourselves, that would be wonderful. Enclosing diagrams of
both datasets, SQL code for mart construction, and the database schema
just in case (and if you need any sample data, please let me know).
My specific question for this use case is: If you search Reagents
dataset, how can you filter by packet ID (a field in packets_tbl)?
Thank you so much for your help.
Best regards,
Marina.
Hi Marina,
during 'naive' MEditor creates attributes from all the columns in all
tables
(ignoring the ones with a '_key' suffix)
and filters from all the columns in the _main_ tables only. The
dimensions
are ignored for filters because, on big datasets they tend to perform
poorly.
However, if you still want them, you can bypass it be creating your
filter 'manually' using
'insert filter' option and editing relevant fields
a.
------------------------------------------------------------------------
-------
Arek Kasprzyk
EMBL-European Bioinformatics Institute.
Wellcome Trust Genome Campus, Hinxton,
Cambridge CB10 1SD, UK.
Tel: +44-(0)1223-494606
Fax: +44-(0)1223-494468
------------------------------------------------------------------------
-------
