Currently, when importing multi-channel images into OMERO, the default rendering model depends on the number of channels. The logic (specifically in RenderingSettingsImpl.resetDefaults using the Renderer.MAX_CHANNELS constant here) behaves as follows:
- 1 channel = Greyscale
- 2-7 channels = RGB/Colour
- 8+ channels = Greyscale
With the rise of highly multiplexed spatial biology instruments, it is becoming increasingly common for users to import entire datasets of images with 10+ channels. Under the current behaviour, all of these images default to greyscale.
Having 8+ channel images default to greyscale creates a confusing initial user experience and possibly leading to a tedious workflow, either manually switch the rendering mode from greyscale to colour image-by-image, or using "Save to all" after disabling greyscale.
The current behaviour can be reproduced by importing a series of fake images with a growing number of channels:
- import&sizeX=4000&sizeY=4000&resolutions=4&sizeC=5.fake (Defaults to Colour)
- import&sizeX=4000&sizeY=4000&resolutions=4&sizeC=7.fake (Defaults to Colour)
- import&sizeX=4000&sizeY=4000&resolutions=4&sizeC=8.fake (Defaults to Greyscale)
A more realistic example can be found in https://zenodo.org/records/7916784 and https://zenodo.org/records/10963121
We propose updating the default rendering behaviour to match the reality of modern multiplexed data. Specifically:
- Remove the 8-channel upper limit for colour rendering such that any image with 2 or more channels should default to the Colour rendering model, regardless of how many channels it has.
- Keep the default behaviour of limiting the active channels to 3 by default when a multi-channel image (2+ channels) is imported.
Users should still be able to manually activate additional channels up to the client limit (default 10 channels in OMERO.web / OMERO.iviewer) after import, but the initial state will be natively rendered in colour, saving on any manual configuration to get there.
Currently, when importing multi-channel images into OMERO, the default rendering model depends on the number of channels. The logic (specifically in RenderingSettingsImpl.resetDefaults using the Renderer.MAX_CHANNELS constant here) behaves as follows:
With the rise of highly multiplexed spatial biology instruments, it is becoming increasingly common for users to import entire datasets of images with 10+ channels. Under the current behaviour, all of these images default to greyscale.
Having 8+ channel images default to greyscale creates a confusing initial user experience and possibly leading to a tedious workflow, either manually switch the rendering mode from greyscale to colour image-by-image, or using "Save to all" after disabling greyscale.
The current behaviour can be reproduced by importing a series of fake images with a growing number of channels:
A more realistic example can be found in https://zenodo.org/records/7916784 and https://zenodo.org/records/10963121
We propose updating the default rendering behaviour to match the reality of modern multiplexed data. Specifically:
Users should still be able to manually activate additional channels up to the client limit (default 10 channels in OMERO.web / OMERO.iviewer) after import, but the initial state will be natively rendered in colour, saving on any manual configuration to get there.