Nuclei texture


I need to measure the nuclei texture after different treatments. In my pipeline I have identified the primary objects, followed by texture measurement. The spreadsheet contains many different measurements related to texture. Which one should I use? I saw in one paper they used DAPI pixel correlation but I can’t find it in the results.

Thank you in advance.


Which texture measure to use is complicated! Because texture can be vague, or at least might mean different things to different people.
My two quick answers are:
(1) Please post a representative picture of each condition, and maybe we can suggest the right measure
(2) Try CellProfiler Analyst’s machine learning Classifier and let it tell you the features that most distinguish your classes!

Now (2) is more complicated, unless you know how to use CellProfiler Analyst (Manual here). But it can be a powerful and informative technique.


Hi David,

Thank you. I have attached an image of each treatment. I will also try to classify the images using CellAnalyst.

Thank you for your help.

The texture differences are apparent, though somewhat subtle. A CPA “feature discovery” approach might work well as I mentioned before. But you could also do simple statistical difference tests (t-tests, or more likely non-parametric tests are needed) on each feature between conditions to discover the greatest differences. This testing on all features would lead to multiple comparison issues if you were using them to combine into a statistical difference test, but if you use this discovery technique on these data, and then do the actual comparison on your discovered separating features from a different set of images/data, then I think you are safe (though consult your local statistician!).

Hope that helps,

Could you please share your pipeline?