Any ideas on how to quantify a non-nuclear immunofluorescent signal in adipocytes? I had the idea to do a watershed on inverted image to "locate" the center of the empty space where lipid droplets used to be, and then maybe estimation would be possible by calculating the AUC of the "first peak" for the plot profiles extending form the "point-maximum" outwards. In case it's not clear, I'll attach an example of what I had in mind with a schematic and what the output would look like in a single inverted HE adipocyte (the same should work for IF). Is there a package in ImageJ/Fiji that could help with this or maybe another way this could be automatized? Somebody working with adipocytes might know? I know there are some packages and even stand-alone stuff for adipocyte morphometry (AdipoCount?) sop for those working with these, do you know if it could be done in any of the software available or has to be custom-made?

Also, this is what first comes to my mind, but there are probably better ways of estimating the cytoplasmic/membrane expression as this provides the output based on rather small sample of the membrane (6 profiles) with each sample suffering from bias based on expression signal from surrounding adipocytes, although I guess this is not extremely important as you are usually interested in them as a group rather than in the expression in single cells. But if you have other ideas I'd be really interested..

Thanks,

jan