way to look for close proximity?

[Rainer Kohler-2]

>Hi,


we are imaging two fluorescent markers that are localized in
different cells but are present in the same tissue according to our
theory. The data consists of z-stacks of images through the tissue.
Is there a mathematical way to analyze and support the close
proximity in the same tissue?
I guess colocalization studies should not give any colocalization.
Can it be called "colocalization" in the same tissue or cooccurence
in the same tissue?

Rainer







--
Rainer Kohler, Ph.D
Microscopist
Center for Molecular Imaging Research
Massachusetts General Hospital/ Harvard Medical School
Building 149, 13th  Street, Room 5406
Charlestown, MA 02129-2060
Tel. [phone]
[hidden email]





The information transmitted in this electronic communication is intended only for the person or entity to whom it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you received this information in error, please contact the Compliance HelpLine at 800-856-1983 and properly dispose of this information.

[Gareth Edwards-3]

Hello,
Voronoi or Delaunay tessellations may be of help, there is a plugin by Johannes Schindelin available:

http://wbgn013.biozentrum.uni-wuerzburg.de/ImageJ/delaunay.html 

Matos et al. (2002) Cell Motility and the Cytoskeleton 53: 53-65 may be a very useful reference as well.

Hope that this is of some help.

Best regards,

Gareth Edwards.


>>> Rainer Kohler <[hidden email]> 05/18/07 4:01 pm >>>
>Hi,


we are imaging two fluorescent markers that are localized in
different cells but are present in the same tissue according to our
theory. The data consists of z-stacks of images through the tissue.
Is there a mathematical way to analyze and support the close
proximity in the same tissue?
I guess colocalization studies should not give any colocalization.
Can it be called "colocalization" in the same tissue or cooccurence
in the same tissue?

Rainer







--
Rainer Kohler, Ph.D
Microscopist
Center for Molecular Imaging Research
Massachusetts General Hospital/ Harvard Medical School
Building 149, 13th  Street, Room 5406
Charlestown, MA 02129-2060
Tel. [phone]
[hidden email]





The information transmitted in this electronic communication is intended only for the person or entity to whom it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you received this information in error, please contact the Compliance HelpLine at 800-856-1983 and properly dispose of this information.

[Martin Wessendorf]

Rainer Kohler wrote:

> we are imaging two fluorescent markers that are localized in different
> cells but are present in the same tissue according to our theory. The
> data consists of z-stacks of images through the tissue.
> Is there a mathematical way to analyze and support the close proximity
> in the same tissue?
> I guess colocalization studies should not give any colocalization.
> Can it be called "colocalization" in the same tissue or cooccurence in
> the same tissue?

My guess is that there are stereological approaches to this problem.
The probability in a random section that object "x" will appear within
distance "k" of object "y" will be a function of how far apart the
objects actually are.  However, I can't point you to a specific paper.
You might check with your local stereology guru or scan through recent
copies of J. Microscopy.

Also, B. Payne and A. Toga published a paper on "distance fields" in
Neuroimaging back around 1992 or so--it's a means of finding objects
within distance "k" of each other.

Finally, there's the ten-cent solution:

Colocalization studies implicitly define a distance within which the two
labels must occur.  For LM studies this is typically either defined by
the cell membrane, or (when one's interested in protein-protein
interactions), the resolution of the microscope.  However, there's no
reason this size couldn't be bigger.

Suppose you resampled your image so that your voxels were, say, 10 um
cubes, and you didn't see any colocalization.  Now suppose you increased
the voxel size to 20um and you did see colocalization.  Those
observations might allow you to set limits on the closeness of the labels.

You'd need strong labeling (or thresholded images) but it might work....

Good luck!

Martin Wessendorf
--
Martin Wessendorf, Ph.D.                   office: [phone]
Assoc Prof, Dept Neuroscience                 lab: [phone]
University of Minnesota             Preferred FAX: [phone]
6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
Minneapolis, MN  55455             E-mail: martinw[at]med.umn.edu