If there is a high-value cytoplasmic intensity around the nucleus, this cell is likely to be normal, and that there was some error in CellPose. To identify dividing cells and cells with a nuclear-only signal, the Nucleus_2 subset is expanded by 5 pixels to check the surrounding intensity of the nuclei. Optional step 3_part 2: Subtract the Nucleus_2 subset from the Nucleus_raw subset to obtain the Leftover_nucleus, a set of nuclei that do not have a CellPose mask associated with it. It is not used for data analysis, but provides an easy way to visualise the object tracking tool downstream of the CellProfiler analysis tool. Called Nucleus_2_fortracking, this generates microscopy images with the nucleus and the CellPose cell mask boundary. Optional step 4: Repeat of the step above, but the smaller object is shrunk prior to subtraction. IdentifyTertiaryObjects function identifies the subset of the original nuclei objects that have a CellPose mask associated with them (Nucleus_2). The subsequent steps allow including cells that have nuclear-only signal.
However, successfully tracking cells across division requires that such cells are included in the analysis. dividing cells) often do not generate a CellPose cell mask and are therefore lost. Optional step 3_part 1: Cells with little cytoplasmic signal (e.g.
Generates a cytoplasm mask by subtracting the nucleus from the CellPose cell masks. This steps relabels each nucleus and corresponding cell mask with the same object number. Since the nuclear and cell objects are identified by two different softwares (CellPose and CellProfiler), the cell and its corresponding nucleus are labelled with a different number. Optional step 2: Manually correction of nuclear CellProfiler segmentation errors. Identifies individual objects as “Nucleus_raw” using the modified SiR-DNA images
Optional step 1: Manual correction of CellPose segmentation errors. Groups the input images by frames to facilitate extracting downstream. tiff file contains timepoints to facilitate object tracking.ġ)Modified sir-DNA images are called ‘Edited_DAPI’Ģ)Original SiR-DNA and KTR images are called ‘Original_DAPI’ and ‘KTR’ respectivelyģ)Cell objects generated from CellPose are converted to CellProfiler objects called ‘Cell_mask_raw’ For live imaging experiments, indicate stacked. tiff file for each channel of live imaging experiment This pipeline quires:ġ) Modified SiR-DNA images from Fiji (for nuclear segmentation)Ģ) Cell objects generated from CellPose (for cytoplasmic segmentation)ģ) Original SiR-DNA and KTR images to measure intentisy etc. The different stages of the "CellProfier_pipeline.cpproj" pipeline, shown in the white tab on the left, are explained below: Stage Below is a summary of the main stages in the pipeline, expected output and optimisation. We used CellProfiler to generate the nuclear and cytoplasmic segmentation, measure the properties of interest, and object tracking for live cell imaging.