Hi John

Obviously, subtracting a single number is not correct.  If you know what 
the background is due to you can subtract a model for its behavior (e.g. 
if it is reflection it should follow the z-response of the microscope) . 
If you have no knowledge of the backgound behavior you can directly 
measure the background in a region where you _know_ there is no signal 
-this could be another z-stack in a region without stained cells. If you 
have a source of "non specific" signal in  the specimen (e.g. cell 
autofluorescence),  you have a bigger problem as that can't be simply 
subtracted -it's actually data. This will increase the number of 
measurements needed to detect a difference between control and treated 
cells for example.

If it's a reflection problem can't you fix it (e.g better filter 
combinations and/or a polarizer)?

My 2c

Cheers Mark Cannell





John Oreopoulos wrote:
> I have some confocal z-stacks of cells expressing a fluorescent 
> protein and I've encountered something I haven't had to deal with 
> before. I'm wondering what is the best way to subtract the background 
> signal from a z-stack like this. My problem is that the background is 
> different for each z-step, peaking a the z-height corresponding to the 
> bottom of the cells that are in contact with the substrate - 
> supposedly since this is the height where you have the most reflection 
> of the excitation light from the substrate itself backwards to the 
> detector. 
>
> Which is better? Should I subtract the maximum background level of the 
> substrate slice from all slices or subtract the background of each 
> individual slice from themselves only? Or I could subtract the average 
> background over all slices from each slice instead of the maximum 
> value at the substrate slice. Alternatively, should I create the 
> z-projection and subtract the background of the z-projection image?
>
> Thanks for anyone who can clear this up for me.
>
>
> John Oreopoulos, BSc,
>
> PhD Candidate
>
> University of Toronto
>
> Institute For Biomaterials and Biomedical Engineering
>
> Centre For Studies in Molecular Imaging
>
>