Is there a way to calculate Manders coefficient with the current modules? If not, it might be a nice feature to add to the Measure Correlation module.
We don’t have it yet, but it’s been on our to-do list for a while. Stay tuned!
Last response to this inquiry was more than two years ago, so I thought I’d revive it. My present data set seems to be well-represented by Manders coefficients (computed in imagej one-by-one. Is it yet possible with CP?
Not yet built in to CP (github.com/CellProfiler/CellProfiler/issues/197) and no timeline on that. However, you can run ImageJ plugins in CP using the RunImageJ module. The easiest way to do this is usually to save a Macro in ImageJ or FIJI and then plug that into RunImageJ.
Thanks David! That seems like a viable work-around. I’ll try it. Keep up the good work, CP developers!
did you found a solution for integrating the Manders coefficient into Cell Profiler. I tried to use the ImageJ Plugin for it, but somehow it is not working yet.
No, we have not attacked Manders coef yet. Our next patch release is imminent and our software engineer is focused on that. Feel free to post your pipeline attempt using RunImageJ, in addition to the ImageJ plugin (or its location at least) so we can try it out and maybe offer some pointers on this workaround.
I too have struggled to find a way to calculate Manders using CellProfiler but, correct me if I’m wrong, I think the value for Precision in CalculateImageOverlap can be used as a Mander’s Coefficient. While it doesn’t take into account the intensities of each pixel, Precision calculated from user defined thresholds converted into binary images should give you an accurate number that represents the percentage of an image that is colocalized with another image.
It is good to point out CalculateImageOverlap, thanks. Without delving into it more we won’t claim any equivalence between Manders and it, but it may satisfy some of you until we determine whether Manders is something we want to add specifically.