Hi, On Tue, 6 Mar 2007, E.O. Stinson wrote:
> Hi, > We at VectorBase are basically in a position where we'll be hosting > several organisms based on data coming out of the Ensembl pipeline. > Currently, that means A. gambiae and A. aegypti. We actually transform > the data into the Chado schema (using the Ensembl Perl API) and then add > some additional data. Once all the data is integrated, we export for use > with Mart. The reason that I've been exploring partitioning is because > creating identical datasets seems like doing it the long way round, but > if partitioning doesn't work yet... We are halfway there to producing this functionality in MartEditor where you can have a single template configuration for all the species but it is very nasty to try and use at the moment. If you have only two species currently I would definitely recommend to sticking with the old route of a separate XML for each. The new MartEditor with templating should hopefully be out soon. > > Is there at least a way to fetch, modify, and then re-upload the Mart > schemas in XML format to allow easier search-and-replace of organism > names? It seems that the Open and Save menu items are missing from the > 0.5 MEditor. > the best option will be to use the Save All option in the File menu, do your text editing on these files and then use the Upload All option let me know how you get on best wishes Damian > --E.O. Stinson > > Arek Kasprzyk wrote: > > > > On 5 Mar 2007, at 22:19, E.O. Stinson wrote: > > > >> Hi all, > >> > >> I've been trying to get dataset partitioning working and it seems to > >> be broken in the release of MartEditor that's publicly available. (For > >> instance, there is no way to add partitions, although I've modified > >> the code I have to get around that.) Now I'm realizing that it's not > >> actually creating the dataset partitions in the database. Instead, > >> it's creating a single dataset with the name of the template, rather > >> than several based on the partition names. Is there a trick to this > >> I'm missing, or a version of MartEditor that does this better? Should > >> I be trying MartBuilder instead? Alternatively, is the other way to do > >> this to create separate datasets for every organism? If so, is there a > >> way to save and load individual XML files? Those seem to have > >> disappeared since version 0.4. > >> > >> For reference, I'm using the release-0_5 CVS branch of MartEditor. > >> > > > > Hi, > > I am afraid partitioning is not yet properly supported. This feature > > will require changes to MBuilder and substantial changes > > to the future releases of MEditor. However, could you give us more > > details on what you are trying to achieve? It is likely > > that we can find the way round it as we of course are in the same > > position and need to add ensembl species quite often. > > > > Are you working with existing XML? eg ensembl? if yes, what version of > > Ensembl is that? > > Are you simply trying to add new species? or you want to do something > > more complicated? > > if so, could you give us more details? > > > > > > thanks, > > a. > > > > > > a. > > > > > >> Thanks, > >> E.O. Stinson > >> > > > > > > ------------------------------------------------------------------------------- > > > > > > 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 > > ------------------------------------------------------------------------------- > > > > > > > > > > >
