[ome-devel] on omero, pytables and genotypes

[Gianluigi Zanetti]

Hi Josh.

On Wed, 2011-02-02 at 17:21 +0100, Josh Moore wrote:
> Since N >> n_recs, a rotated table is one option for using OMERO right
> away, with n_recs*N rows with vid & act as denormalized values
> repeated in each column, but I didn't go through the exercise of
> seeing what that does to your computations. Just FYI:
> 
> 

Well, part of the problem is that each row, a column in your rotated
scheme, correspond to the genotyping of a specific sample, and it is
what you keep accumulating (e.g., we have about 10k of them and more
will come). Thus, while N, the number of SNPs is fixed by the specific
technology, the number of records keeps increasing. 
A possible alternative is to do some fromstring/tostring magic.
We will probably try that as a temporary solution.
It would be nice if we could have direct numpy array columns a la
pytable, however, because then it would be possible to explicitly do
column selections server side. 

--gianluigi