How to create a plate with multiple images per well

[Will]

I'm new to OMERO and am having trouble understanding the best way to create a plate with multiple images per well. For example, if I import a dataset containing 24 images corresponding to a plate size of 3 columns by 4 rows, with 2 images per well, how does one represent 2 images per well when creating a plate using the Dataset to Plate script? One possible solution is to use image files containing z-stacks. Is this the preferred solution to this problem in OMERO?

[wmoore]

Hi,

I've updated the Dataset_To_Plate script to support multiple Wells Per Image.
PR opened at https://github.com/ome/scripts/pull/146 (not yet tested by the team).

Hope that does what you want?

Regards,
Will.

[Will]

Thanks for the new script Will. I tested it and it seems to work just fine. I am a software developer and I showed it to the scientist on our team that I'm working with. The scientist was wondering if this is the correct way to populate a plate in omero if the images in question are separate "channels" as opposed to separate "fields" in a well. The script in question seems to create a group of fields per well as opposed to a group of channels.

Question 1:
Is there a different procedure for loading plates with multiple channel images per well as opposed to multiple field images per well? Or, is the same procedure used for both?

Question 2:
Does the new script you provided load the images in file name order. For example, if I have img_001.tif, img_002.tif, img_003.tif, etc, I think the script will load the images in order 1, 2, 3. This means that each image field or channel must be in order. For example, assuming 3 images per well, the following order must be maintained:

img_001.tif -> well A1 channel 1
img_002.tif -> well A1 channel 2
img_003.tif -> well A1 channel 3
img_004.tif -> well A2 channel 1
img_005.tif -> well A2 channel 2
img_006.tif -> well As channel 2
...

Is this the correct interpretation?

[wmoore]

Hi,

The Dataset_To_Plate script assumes that you already have multi-dimensional images (doesn't attempt to combine channels etc).

There is a Combine_Images script in the same util location that can take multiple Datasets and combine the single-plane images in each Dataset into a new image (per Dataset).
See "Combine Images" under https://help.openmicroscopy.org/scripts.html#utility

However, Combine_Imaged duplicates the data. A better way to do this is to combine the images at import time. If you are importing a bunch of e.g. Tiffs or Pngs, you can first create a pattern file See https://docs.openmicroscopy.org/bio-for ... -file.html
This should allow you to create multi-channel images on import. Then you can use the Dataset_To_Plate script to assemble these into Plates.

[Will]

Thanks for the info Will. Is the capability of grouping files with a pattern file available in OMERO 5.3.5? I don't see that manual page in the 5.3.5 documentation. I have not tried it yet but now I'm wondering if it's even available in my version.

[sbesson]

Hi Will,

I don't see that manual page in the 5.3.5 documentation

This is primarily because from the OMERO standpoint, pattern files are no different from any other format supported by Bio-Formats.

Is the capability of grouping files with a pattern file available in OMERO 5.3.5?

Support for file patterns was initially introduced in Bio-Formats 5.1.0 and OMERO 5.3.5 includes Bio-Formats 5.5.3. Some improvements might have been added since but I expect the main functionalities should be available in your installation.

Let us know how this goes. Best,
Sebastien

[Will]

I'm a little confused about grouping files using a file pattern as described here: https://docs.openmicroscopy.org/bio-for ... -file.html. I tried a test using 6 files named as follows:

test_z0_c0.tiff
test_z1_c0.tiff
test_z2_c0.tiff
test_z0_c1.tiff
test_z1_c1.tiff
test_z2_c1.tiff
test_z0_c2.tiff
test_z1_c2.tiff
test_z2_c2.tiff
rgb.pattern

The rgb.patter

The test.pattern file had the following contents:
test_z<0,1,2>_c<0,1,2>.tiff

I imported the rgb.pattern file but the result was a single tiff file containing 3 Z layers. I was expecting to get 3 tiff files containing 3 Z layers each. What am I doing wrong or misunderstanding in the import process?

[sbesson]

Hi,

assuming each of your TIFF file contains single planes of the same dimensions, then the following pattern:

test_z<0,1,2>_c<0,1,2>.tiff

should represent a single multi-dimensional image with 3 channels and 3 focal planes.

I was expecting to get 3 tiff files containing 3 Z layers each.

If you are expecting to represent your data as 3 separate images with 3 Z layers each, you might need to create three separate pattern files e.g.

Code: Select all

test_z<0,1,2>_c0.tiff


Best,
Sebastien