Analysis of Multidimensional Microscopy Data Using Cell-ACDC

We are improving the analysis of multidimensional microscopy data by developing a software for analysis of cell division cycle, named Cell-ACDC, to overcome the bottlenecks to fast biological discovery. Current AI models are often hard to access. Additionally, visualization and manual correction are required for achieving high-quality results.

These tasks, however, can become very tedious without the right tools. To begin, click on Launch GUI in the main window of the visualize and correct module. Click the folder icon in the toolbar of the new window, and select the folder containing the data.

Then press Select Folder to confirm the selection. Use the dropdown menu to select the channel phase contrast pre-processed, then press OK to confirm. Select the segmentation mask name and then click Load Selected to load the segmentation file created in the previous step.

Confirm the image properties by clicking OK for loaded Positions. When prompted, select No to prevent loading of additional fluorescence data. Use the mode selector to select segmentation and tracking mode.

In the menu bar, navigate to Tracking, followed by Select real-time tracking algorithm, and select the desired real-time tracker based on the organism. Use the left and right arrow keys to navigate between frames. Navigate to frame 10.

Press the key S to activate the manual bud separation tool and right-click to automatically split the segmentation mask of cell one. Now, navigate to frame 14. Press the key B to activate the brush tool and draw the missing segmentation mask for the bud using the left mouse button.

Continue through the subsequent frames while correcting segmentation and tracking errors using the available tools. Correct at least until frame 42. Activate cell cycle analysis using the mode selector.

When prompted, select Yes to go to frame one. Use the left and right arrow keys to navigate between frames. Click OK to accept the initialization of the cell cycle annotation table when prompted, and navigate to frame 41.

Right-click on cell one or its bud to separate the connection and annotate the cell division event. Continue viewing all relevant frames and correct any mistakes in automatic mother-bud assignments using the available tools. To assign a bud to a mother, activate the assigned bud to mother tool by pressing A.Press and hold the right mouse button on the bud, drag to the corresponding mother cell, and release the mouse button.

To reinitialize the cell cycle annotation, select the appropriate option from the toolbar. To break or rebind the mother-bud association, ensure that no tool is selected. Right-click on an existing mother-bud pair to break the association, or right-click again to reestablish the connection.

Activate normal division lineage tree using the mode selector. When prompted, select Yes to go to frame one, and use the left and right arrow keys to navigate between frames. Correct errors in the automatic mother-daughter assignments using the tools available in the Edit toolbar.

When prompted, click Propagate to apply the changes. To assign a mother to a new cell ID, activate the find mother for a new cell ID tool by pressing F.Right-click on the new cell to cycle through candidate mothers. Nuclear segmentation in tumor spheroids revealed a wide distribution of nucleus volumes, with a substantial number of objects displaying small volumes and a few showing extremely large volumes.

A 3D view of the tumor organoid displayed numerous segmented nuclei with labeled identifiers, and z-slices showed the red segmentation contours applied to each nucleus. In budding yeast, the H2B protein amount increased sharply at the time of bud emergence and plateaued prior to nuclear division. The number of nuclei increased suddenly at the time of nuclear division in the yeast dataset.

In mouse embryonic stem cells, the cell area increased steadily until reaching a maximum, then decreased during cell division, and later began to rise again in the daughter cells. So Cell-ACDC is an open-source software framework that enables easy access to AI models for bioimage analysis and ensures high shareability of microscopy data. An important aspect of Cell-ACDC is that the community can easily integrate the new methods into an existing workflow with standardized data structure.

Leveraging corrected data from Cell-ACDC for fine-tuning state-of-the art methods can lay the foundation for fully automated bioimage analysis.