Examples of worklfows combining Fiji and OMERO

Environment Setup

The specific requirements for each macro are listed in the corresponding sections.

Workflows

FixSizeDownloader.groovy

The macro is dervived from modifying the following script: https://github.com/ome/omero-guide-fiji/blob/master/scripts/groovy/open_image_after_download.groovy

The macro is at: https://github.com/German-BioImaging/fiji_omero_workflows/blob/main/macros/FixSizeDownloader.groovy

This macro is designed to download large images from OMERO and then open them at a lower resolution in Fiji, as the current plugins would only open the image at the highest possible resolution. The macro:

1. Connects to OMERO.
2. Downloads the fileset (can consist of multiple file associated to the image in OMERO).
3. Open the files at the chosen pyramid level.
4. Correct the pixel width and pixel height (match the resolution of the pyramid level).
5. Delete the downloaded files.

Input

From a dialog window after starting the macro:

• Username = OMERO username.
• Password = OMERO password.
• Host = Address of the OMERO server.
• Port = Port of the OMERO server (default = 4064).
• Image ID = ID of the image (needs to be in the default group of the user, TO DO: Add group switching)

Output

No output, the downloaded image is deleted. The image with the chosen level of resolution is opened in Fiji, but not saved locally.

CountCellsOMERO.ijm

The macro is at: https://github.com/German-BioImaging/fiji_omero_workflows/blob/main/macros/CountCellsOMERO.ijm

This macro processes all images in a given dataset (or multiple datasets with a given tag) and measures cell numbers in each ROI matching a given prefix. Cells touching the ROI border are discarded. Images with no matching ROIs are skipped. The resulting ROIs are saved back to OMERO, togheter with tables reporting the number of cells.

Additional requirements

• OMERO_macro-extensions (https://github.com/GReD-Clermont/omero_macro-extensions)
• ColorDeconvolution2 (https://blog.bham.ac.uk/intellimic/g-landini-software/colour-deconvolution-2/)
• StarDist and CSBDDeep plugins

Input

From a dialog window after starting the macro:

• Username = OMERO username.
• Password = OMERO password.
• Host = Address of the OMERO server.
• Port = Port of the OMERO server (default = 4064).
• Group = ID of the group the dataset belongs to (default = 0, keeps the user's default group).
• Process By Tag = Select multiple datasets to process (default = false)
• Tag Name = All datasets with these tags in the current group are processed (only if Process By Tag = True) (default = 'to_process').
• Dataset ID = OMERO id of the dataset to process (only considered when Process By Tag = false can be looked up from OMERO.web).
• ROI prefix = Prefix of the ROI names to analyze, only matching ROIs are processed.
• StarDist Model = Name of the StarDist model to use (default = 'Versatile (fluorescent nuclei)')).
• StarDist Normalize Input (true/false)" = StarDist normalizeInput parameter (default = 'true'`)
• StarDist percentileBottom = StarDist percentileBottom parameter (default = 25)
• StarDist percentileTop = StarDist percentileTop parameter (default = 100)
• StarDist probThresh = probThresh StarDist parameter (default = 0.4)
• StarDist nmsThresh = nmsThresh StarDist parameter (default = 0.4)

Output

The following tables are saved to OMERO:

• Attached to each image: DATE_TIME_CellCount_IMAGEid
• Attached to the dataset: DATE_TIME_CellCountSummary_DATASETid

The tables contain the following columns:

• Image = OMERO image id.
• Name = Name of the ROI.
• X = X location of the top-left corner of the ROI.
• Width = Width of the ROI.
• Y = Y location of the top-left corner of the ROI.
• Height = Height of the ROI.
• C = c slice.
• Z = z slice.
• T = t slice.
• Label = OMERO image id.
• ImageName = Image name.
• Total_area_um2 = Total Area of the bounding box of the ROI (µm^2 if the pixel unit is µm).
• Total_ROI_area_um2 = Total Area of the ROI (µm^2 if the pixel unit is µm).
• CellCount = Number of segmented cells.

Additionally, ROIs for all segmented cells are added to each image.

Attribution

The macro was developed by Michael Gerlach for Anett Jannasch and later adapted to work with OMERO by Tom Boissonet and Michele Bortolomeazzi.

The context this macro was originally applied in, togheter with its original version can be found at: https://doi.org/10.1063/5.0182672

TargetQuantificationOMERO.ijm

The macro is at: https://github.com/German-BioImaging/fiji_omero_workflows/blob/main/macros/TargetQuantificationOMERO.ijm

This macro processes all images in a given dataset (or multiple datasets with a given tag) and measures the area positive for collagen or elastin (see code for color deconvolution and thresholding parameters). Images with no matching ROIs are skipped. The measured areas are saved as tables in OMERO.

Additional equirements

• ColorDeconvolution 2 (https://blog.bham.ac.uk/intellimic/g-landini-software/colour-deconvolution-2/)

Input

From a dialog window after starting the macro:

• Username = OMERO username.
• Password = OMERO password.
• Host = Address of the OMERO server.
• Port = Port of the OMERO server (default = 4064).
• Group = ID of the group the dataset belongs to (default = 0, keeps the user's default group).
• Process By Tag = Select multiple datasets to process (default = false)
• Tag Name = All datasets with these tags in the current group are processed (only if Process By Tag = True) (default = 'to_process').
• Dataset ID = OMERO id of the dataset to process (only considered when Process By Tag = false can be looked up from OMERO.web).
• Target Molecule (collagen or elastin) = Target molecule (default = 'collagen').
• ROI prefix = Prefix of the ROI names to analyze, only matching ROIs are processed.

Output

The following tables are saved to OMERO (the same unit as the pixel size is used):

• Attached to each image: DATE_TIME_TARGET_IMAGEid
• Attached to the dataset: DATE_TIME_TARGET_Summary_DATASETid

The tables contain the following columns:

• Image = OMERO image id.
• Name = Name of the ROI.
• X = X location of the top-left corner of the ROI.
• Width = Width of the ROI.
• Y = Y location of the top-left corner of the ROI.
• Height = Height of the ROI.
• C = c slice.
• Z = z slice.
• T = t slice.
• Label = OMERO image id.
• ImageName = Image name.
• Total_area_um2 = Total Area of the bounding box of the ROI (µm^2 if the pixel unit is µm).
• Total_ROI_area_um2 = Total Area of the ROI (µm^2 if the pixel unit is µm).
• Total_Positive_area_um2 = Total area covered by the target moleclue in the ROI (µm^2 if the pixel unit is µm).
• Fractional_area_percent = Fractional (target moleule positive area / ROI area) area as a percentage.

Attribution

The macro was developed by Michael Gerlach for Anett Jannasch and later adapted to work with OMERO by Tom Boissonet and Michele Bortolomeazzi.

The context this macro was originally applied in, togheter with its original version can be found at: https://doi.org/10.1063/5.0182672