I want to have CellProfiler calculate the mitotic index of a series of images by classifying cells as phospho-H3 positive or negative. I know how to identify nuclei and measure their phospho-H3 levels with MeasureObjectIntensity; what do I do next?
You can use the ClassifyObjects module to determine the percentage of objects above a threshold.
First, choose a measurement to use that determines if your cells are positive or negative, in this case, either Integrated Intensity or Mean Intensity of PH3 staining.
Then, where it asks for the number of bins, you can type P:XXX where XXX is the numerical cutoff you want to use to determine if a cell is positive or negative. This can be determined by first processing a few sample images and using the data tool ShowDataOnImage to visually see the measurement values for different cells.
After you have chosen a cutoff, make sure to use a lower limit of 0 and an upper limit which is very high such as 999999 so you do not exclude any cells. Or, if you have very bright auto-fluorescent cells, you can determine their measurement value and use something lower than that to exclude them from the classification.
That’s it! You can then choose to save the result as an absolute value (number of cells) or a percent value (percent of cells out of total).
One more question on this topic -
I want to classify cells as either mitotic or interphase cells based on phospho-H3 staining. Then I want to run later modules on each category. When I use the ClassifyObjects module, it creates a category of metaphase cells defined as the nuclei with PH3 staining above a threshhold - that works fine. Now I want to define a category of interphase cells as nuclei with PH3 staining below a threshhold. The ClassifyObjects module only seems to let me define a category as objects above a threshhold. What’s the best way to create a category of interphase cells?
Never mind, I just figured it out. FilterByOBjectMeasurement lets me define categories either above or below a threshhold.