Are "labels" supported by the plugin?

[constantinpape]

I have an ome.zarr file with a multi-channel image and two label images, a cell segmentation and a nucleus segmentation:

$ tree -L 2 MMStack_Pos0.ome.zarr
MMStack_Pos0.ome.zarr
├── 0
...
└── labels
    ├── cells
    └── nuclei

When I open this file via the napari-ome-zarr plugin the image is opened and displayed correctly, but the labels are not opened:

napari-ome-zarr.mp4

Are "labels" supported by the plugin? (If yes, then maybe something with how I save the labels is incorrect.)
The example file is available here: https://oc.embl.de/index.php/s/ezHha9qb8HjxSfd

[constantinpape]

cc @sbesson

[joshmoore]

ome_zarr -vvv info ... output can be useful with these types of issues.

[constantinpape]

Here's the output I get via info:

$ ome_zarr -vvv info MMStack_Pos0.ome.zarr
DEBUG:ome_zarr.format:Created nested FSStore(MMStack_Pos0.ome.zarr, r, {'dimension_separator': '/', 'normalize_keys': False})
DEBUG:ome_zarr.format:0.4 matches 0.4?
DEBUG:ome_zarr.reader:treating /home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr [zgroup] as Multiscales
INFO:ome_zarr.reader:root_attr: multiscales
DEBUG:ome_zarr.reader:[{'axes': [{'name': 'c', 'type': 'channel'}, {'name': 'z', 'type': 'space', 'unit': 'micrometer'}, {'name': 'y', 'type': 'space', 'unit': 'micrometer'}, {'name': 'x', 'type': 'space', 'unit': 'micrometer'}], 'datasets': [{'coordinateTransformations': [{'scale': [1.0, 4.0, 0.17, 0.17], 'type': 'scale'}], 'path': '0'}, {'coordinateTransformations': [{'scale': [1.0, 4.0, 0.34, 0.34], 'type': 'scale'}], 'path': '1'}, {'coordinateTransformations': [{'scale': [1.0, 4.0, 0.68, 0.68], 'type': 'scale'}], 'path': '2'}, {'coordinateTransformations': [{'scale': [1.0, 4.0, 1.36, 1.36], 'type': 'scale'}], 'path': '3'}, {'coordinateTransformations': [{'scale': [1.0, 4.0, 2.72, 2.72], 'type': 'scale'}], 'path': '4'}], 'version': '0.4'}]
INFO:ome_zarr.reader:datasets [{'coordinateTransformations': [{'scale': [1.0, 4.0, 0.17, 0.17], 'type': 'scale'}], 'path': '0'}, {'coordinateTransformations': [{'scale': [1.0, 4.0, 0.34, 0.34], 'type': 'scale'}], 'path': '1'}, {'coordinateTransformations': [{'scale': [1.0, 4.0, 0.68, 0.68], 'type': 'scale'}], 'path': '2'}, {'coordinateTransformations': [{'scale': [1.0, 4.0, 1.36, 1.36], 'type': 'scale'}], 'path': '3'}, {'coordinateTransformations': [{'scale': [1.0, 4.0, 2.72, 2.72], 'type': 'scale'}], 'path': '4'}]
INFO:ome_zarr.reader:resolution: 0
INFO:ome_zarr.reader: - shape ('c', 'z', 'y', 'x') = (4, 6, 2048, 2048)
INFO:ome_zarr.reader: - chunks =  ['1', '1', '512', '512']
INFO:ome_zarr.reader: - dtype = uint16
INFO:ome_zarr.reader:resolution: 1
INFO:ome_zarr.reader: - shape ('c', 'z', 'y', 'x') = (4, 6, 1024, 1024)
INFO:ome_zarr.reader: - chunks =  ['1', '1', '512', '512']
INFO:ome_zarr.reader: - dtype = float64
INFO:ome_zarr.reader:resolution: 2
INFO:ome_zarr.reader: - shape ('c', 'z', 'y', 'x') = (4, 6, 512, 512)
INFO:ome_zarr.reader: - chunks =  ['1', '1', '512', '512']
INFO:ome_zarr.reader: - dtype = float64
INFO:ome_zarr.reader:resolution: 3
INFO:ome_zarr.reader: - shape ('c', 'z', 'y', 'x') = (4, 6, 256, 256)
INFO:ome_zarr.reader: - chunks =  ['1', '1', '256', '256']
INFO:ome_zarr.reader: - dtype = float64
INFO:ome_zarr.reader:resolution: 4
INFO:ome_zarr.reader: - shape ('c', 'z', 'y', 'x') = (4, 6, 128, 128)
INFO:ome_zarr.reader: - chunks =  ['1', '1', '128', '128']
INFO:ome_zarr.reader: - dtype = float64
DEBUG:ome_zarr.io:open(ZarrLocation(/home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr/labels))
DEBUG:ome_zarr.format:Created nested FSStore(/home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr/labels, r, {'dimension_separator': '/', 'normalize_keys': False})
DEBUG:ome_zarr.reader:treating /home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr [zgroup] as ome-zarr
DEBUG:ome_zarr.reader:returning /home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr [zgroup]
/home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr [zgroup]
 - metadata
   - Multiscales
 - data
   - (4, 6, 2048, 2048)
   - (4, 6, 1024, 1024)
   - (4, 6, 512, 512)
   - (4, 6, 256, 256)
   - (4, 6, 128, 128)
DEBUG:ome_zarr.utils:[dask.array<from-zarr, shape=(4, 6, 2048, 2048), dtype=uint16, chunksize=(1, 1, 512, 512), chunktype=numpy.ndarray>, dask.array<from-zarr, shape=(4, 6, 1024, 1024), dtype=float64, chunksize=(1, 1, 512, 512), chunktype=numpy.ndarray>, dask.array<from-zarr, shape=(4, 6, 512, 512), dtype=float64, chunksize=(1, 1, 512, 512), chunktype=numpy.ndarray>, dask.array<from-zarr, shape=(4, 6, 256, 256), dtype=float64, chunksize=(1, 1, 256, 256), chunktype=numpy.ndarray>, dask.array<from-zarr, shape=(4, 6, 128, 128), dtype=float64, chunksize=(1, 1, 128, 128), chunktype=numpy.ndarray>]
DEBUG:ome_zarr.reader:returning /home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr/labels [zgroup] (hidden)
not an ome-zarr node: /home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr/labels [zgroup] (hidden)

so I guess it does not recognize this as a label node.

For some context: I have implemented the code to write this based on discussions in ome/ome-zarr-py#171. I will post a link to the actual code shortly.

[sbesson]

Fromt the error message

not an ome-zarr node: /home/pape/Work/data/shila/ngff/MMStack_Pos0.ome.zarr/labels [zgroup] (hidden)

I suspect what is missing here is the intermediate labels metadata registering the individual label images

(base) sbesson@Sebastiens-MacBook-Pro ~ % ls -alh ~/Downloads/MMStack_Pos0.ome.zarr/labels 
total 24
drwxr-xr-x@  6 sbesson  staff   192B 15 Mar 16:33 .
drwxr-xr-x@ 11 sbesson  staff   352B 15 Mar 16:33 ..
-rw-r--r--@  1 sbesson  staff   6.0K 15 Mar 16:33 .DS_Store
-rw-r--r--@  1 sbesson  staff    24B 15 Mar 15:20 .zgroup
drwxr-xr-x@  9 sbesson  staff   288B 15 Mar 15:20 cells
drwxr-xr-x@  9 sbesson  staff   288B 15 Mar 15:20 nuclei

At least in my mind, the generation/update of the metadata for this intermediate group should definitely be part of the API discussed in ome/ome-zarr-py#171.

[constantinpape]

@sbesson yes, that's the issue. The labels are found now.
(Though I will have a lot of follow up issues after dinner ;))