Many Channels to 1 Logical Channel

[wmoore]

I noticed that if you import the mias Plate at

from_skyking/mias/frans/siRNA_PRIM1_03102008

then query the DB (after finding most recent logical channel)

omero=# select * from Channel where logicalchannel = 4451;

...
 5002 |   255 |    0 |         -40 |     0 | 255 |         |      160423 |             |        3 |        3 |    160427 |           4451 |   4401 |      4932 |            1
 5001 |   255 |    0 |         -40 |     0 |   0 |         |      160423 |             |        3 |        3 |    160427 |           4451 |   4401 |      4931 |            0
(32 rows)

you find that ALL the Channels (2 channels * 16 images) are linked to one Logical Channel.

Noticed this when renaming channels #10022.

This has implications if you want to have different names for the 2 channels in each image OR for future DB upgrades where Channel & Logical Channel may be merged.

Should probably ensure Channel : LogicalChannel? is 1:1.

[mlinkert]

PR opened: ​https://github.com/openmicroscopy/openmicroscopy/pull/582

[wmoore]

Since creating this ticket, I was looking at how it's handled in other Plates I import: I see that for a whole Plate, the number of Logical Channels is equal to the number of Channels in each image. E.g. in this example, an Image from a Plate has 2 Channels and 2 LogicalChannels?:

omero=# select id, logicalchannel from channel where pixels=3901;
  id  | logicalchannel 
------+----------------
 4252 |           4152
 4251 |           4151
(2 rows)

The problem in the data I imported for this ticket is that there is only 1 logical channel for this Plate, even though every image has 2 channels. E.g:

omero=# select id, logicalchannel from channel where pixels=4417;
  id  | logicalchannel 
------+----------------
 5034 |           4452
 5033 |           4452
(2 rows)

There simply needs to be 2 Logical Channels for this plate instead of 1 (from_skyking/mias/frans/siRNA_PRIM1_03102008).

[mlinkert]

This needs to be checked again to verify that it's still a problem, but I'd expect it to require a fix in OMERO and not Bio-Formats.