photobleaching

[Ma, Limei]

Does anyone have a macro to correct for photobleaching?  When I look at
the stack profile of my images, the photobleaching of GFP appears to
follow an exponential decay curve.  It is OK to process image stack with
by subtracting Z stack min.  But the ideal way is to calculate a pixel
by pixel decay curve from the stack and then do subtraction.  Any
suggestions?

LM

[Brooke Kelley]

WCIF has some nice documentation on the issue:

http://www.uhnresearch.ca/facilities/wcif/imagej/t.htm#t_bleach

You find your decay rate and use the plugin.  Deciding where to calculate
your decay rate from (cell body, process, etc) may be important.  I have
played with it a little and just plotted the mean pixel intensity of the
entire field and calculated a rate from that, but it could be much more
complex.  I decided it wasn't appropriate for what I wanted.  Most
importantly (quoted from the page):

"Since bleaching is often not mono-exponential, quantification of
fluorescence intensities after bleach correction is not possible. This
plugin should really only be used to enhance time-course movies for
presentation rather than quantification."

Brooke Kelley


On Tue, 1 Aug 2006, Ma, Limei wrote:

> Does anyone have a macro to correct for photobleaching?  When I look at
> the stack profile of my images, the photobleaching of GFP appears to
> follow an exponential decay curve.  It is OK to process image stack with
> by subtracting Z stack min.  But the ideal way is to calculate a pixel
> by pixel decay curve from the stack and then do subtraction.  Any
> suggestions?
>
> LM
>

[Michael Cammer]

The better way to do this is to take a few series of bleaching images,
measure the bleach rates of each, and if they really are consistent and
similar, then fit to a curve that then is applied to the images.

The quick and dirty way is the following.  Put a region of interest over a
feature that you believe is changing intensity only because of
bleaching.  Then run.

//-----------------------------------------------
macro 'Multiply to fix bleaching based on ROI'{
   run("Set Measurements...", "  mean  redirect=None decimal=0");
   run("Plot Z-axis Profile");    run("Close");
   standard = getResult("Mean",(0));
   for (i=1; i<=nSlices; i++){
     bgmean = standard / getResult("Mean",(i-1));
     run("Set Slice...", "slice="+i);
     run("Select All");
     run("Multiply...", "slice value="+bgmean);
}

The reason the factor is named "bgmean" is because this was modified from a
background subtraction macro.

-mc


At 04:52 PM 08/01/06 -0500, you wrote:
>Does anyone have a macro to correct for photobleaching?  When I look at
>the stack profile of my images, the photobleaching of GFP appears to
>follow an exponential decay curve.  It is OK to process image stack with
>by subtracting Z stack min.  But the ideal way is to calculate a pixel
>by pixel decay curve from the stack and then do subtraction.  Any
>suggestions?
>
>LM

____________________________________________________________________________
Michael Cammer   Analytical Imaging Facility   Albert Einstein Coll. of Med.
URL:  http://www.aecom.yu.edu/aif/