This tutorial shows, step by step, how to export ImageJ 3D segmentation to MorphoNet. It guides users from the segmented data to final results.
Three dimentional data of live embryonic development at single-cell resolution were taken through light-sheet microscopy as explained in Guignard, Fiuza et al., bioRxiv 238741 (2017).
In this tutorial we take as example the data corresponding to the embryo referred to as A1 in the paper.
Data for this embryo can be found at: http://www.crbm.cnrs.fr/Astec/Datasets/A1.zip. This zip file contains 180 segmented time steps and the associated quantitative information.
You can also directly download the ascidian data associated to all MorphoNet tutorials http://www.morphonet.org/Tutorials/MorphoNet_DATA.zip
Unzip your archive and also the first time point DATA/SEGMENTED/Segmented_t001.tiff.gz file.
Open the .tiff file with Fiji.
Open the menu "Plugins" and search MorphoNet. If the plugin is not visible maybe you need to restart ImageJ/Fiji.
Browse the MorphoNet menu to find "Connection" and click on it. Enter your personnal ID and Password offered to connect to MorphoNet website.
An new box will appear and validate that you are successfully connected.
This connection is dependant of your ImageJ/Fiji session.
If you have to restart ImageJ/Fiji please redo the connection step.
Create a new personnal dataset by clicking on "Create a Dataset".
Write a new Dataset name and validate with the "OK" button.
You can also log to an old dataset. Click on "Pick a Dataset". The older Datasets precedently created have to appear in the dropdown windows.
There is two different treatments proposed by MorphoNet conversion plugin, automatically dependant of the format of the image.
First, you can convert your 3D images in Meshes with the menu "Compute meshes".
You have the possibility to use three different parameters:
Depend on the background and the information you dont want to convert. In this particular case, Background is equal to 1 so you have to enter 1.
Determine the image resolution you want to use to compute the meshes influencing the rendering and the computation time. (If 1 there is no reduction of the size of the image converted and the computation can take 1 hour. With a factor of 4, the time is reduce to few minutes)
Reduce the quality and the size of the Meshes with no major difference in computation time. (Rendering parameter).
After calibration click on "OK" and wait.
An other possibility is to convert your 3D+time dataset in Meshes using the menu "Compute meshes".
For this second case you can try and open the drosophila_nucleus Dataset present in DATA/drosophila_object-identities_000-003.tif.
In addition to the choice of the Threshold (here 0), the resampling factor an the Meshes simplification factor, you can calibrate the begin and the end of the sequence you want to convert. It results in one Meshes image for each time point.
You can follow the advancement of the computation for each cell by the progression of a dot moving in the control windows of ImageJ/Fiji.
After a while your data are converted in Meshes and this is confirmed by a new box.
You can then select "upload meshes" in the MorphoNet menu.
This step is almost instantaneous. And a confirmation box appear. Click on "OK".
A last option is the computation of some measures directly on ImageJ/Fiji.
You can compute measurement using preexisting plugins like "3D Geometrical Measure". You can download the 3D ImageJ Suite for more information .
You will obtain a result table
To upload this measures on Morphonet you can simply click on the "upload result" menu.
The "Value" is the identification of each cells present in the segmenation and is the same used to nammed the Meshes cells.
You can choice which measure you want to upload.
And click on "OK" to validate".