colocalization greater than 100%

[Gretchen Unger, Ph.D.]

Hi all,

I used the division feature of Image J, v. 1.38 to estimate maximum
colocalization between two optical sections, one was green (reporter
gene) and the other red (cells). It looked like it worked but in one
section I calculated a value over 100%. Can anyone explain how that
might happen?

I transformed final images processed against background into
grayscale using Photoshop for the Image J calculations. In Image J, I
calculated a joint (green+red) area by pixel-by-pixel division of
green/red. This would calculate a maximum colocalization as a pixel
with any signal above background in both images would divide to a
"1". Because of drift or jumping or whatever, I had to manually alter
alignment in some of the joint picture calculations.

Next, I calculated "area fractions" in both the joint and red image
using the "threshold" feature, applying threshold , and then
"measuring" following selection of "area fraction" as an attribute.
Because my final calculation was Colocalization =
(green/red)*100/red, I selected the thresholding level from my
denominator using the auto function. I then manually set the
threshold to the same value (whatever it was) for the area fraction
measurement of the joint image (the numerator). I got a number around
170% (17.25 divided by 10.37),  I have used this method in the past
and got numbers ranging from 50% to 98%, which seemed reasonable.  I
don't know why this set of aligned panels exceeds 100% and can't
explain to anyone how that happened or what it can mean. Should I
auto-threshold both images ? did I alter a calculation basis somehow
when I manually aligned the image pair?

Any thoughts or comments would be appreciated.

Gretchen Unger

[Daniel James White]

Hi Gretchen,

your method looks interesting,
but are there are several places where subjectivity creeps in?

To do pixel intensity spatial correlation coloc analysis
try the coloc plugins from macbiophotonics / imageJ plugins pages

First you must subtract background first - maybe use rolling ball or
subtract mean of a background area. You have to decise what is  
background
and what isn't.

The auto threshold (costes method) in the
Colocalization Threshold plugin
will give you a reliable and repeatable non subjective threshold
bases on Pearsons correlation.

It will then calculate the Manders coefficients (amount of red coloc  
with green and vice versa
Manders 1993 i think)
and show you the 2D histogram or scatter plot,
which is a good way to visualise the correlation between the 2 colour  
channels.

I would publish the Manders coefficients, the thresholds that are  
calculated, and the scatter plot (rather than a 2 colour merged image  
- which is very difficult to interpret scientifically )

Then use the colocalisation test plugin,
with the Costes method to test the statistical significance of the  
colo you see.
The coloc result are not solid without this test , as they could be  
caused by random
overlap in a busy image.

Read the coloc docs at macbiophotonics:
http://www.macbiophotonics.ca/imagej/colour_analysis.htm


Other things to consider to get good results.
1) you should image beads on the same system to see any spatial and  
chromatic abberations in the spatial locations of the 2 colour  
channels over the field of view.
You might be surprised how bad it is. (seems you are trying to do that  
already)

2) on a zeiss confocal there are 3 emsission pinholes to adjust with  
100 nm fluorescent beads
to get the signals coregistered. The engineeers only use the 10x lens  
and large beads,
and always leave it in a non optimised state, no good for coloc of  
small objects.

3) Olympus and leica confocals only have a single pinhole,
so als long as you are using a good (expensive) chromatically and  
field corected lens
you are in good shape....

4) ... unless you are using a UV (say 405) laser coming through a  
different optical fiber
into the system than the visible laser lines. The collimator must be  
carefully adjusted to bring the
"blue" signal illuminated by the lioght from the UV fiber into  
register with the visible illumination.

5) 2D images are ok for coloc, but real sampkles are 3D so in order to  
get a good idea of the whole sample 3D imagging is a good thing.

6) If you are using a widefield system, consider deconvolution,
to increase contrast and remove out of focus fuzz,
but thats a whole kettle of fish in itself)

7) You might consider an object based coloc method,
depending on how your biology works.
Segment out objects, then see how much they overlap
between channels.

8) Bleed through of fluorescence from one channel into another is your  
worst enemy in coloc.
You can see it clearly in the 2D histogram.
If you have bleed through, you will likely measure it as colocalisation,
which is in fact false. Watch out!

9) BioImageXD is also free like imageJ ,
and gives the same methods as in the plugins above,
but in a more coherent user interface.
If you have problems with that , send me some sample images,
and we will try to get it to work for you.


Say Hi to MN for me, I used to live in Rochester... Mayo Clinic...


cheers

Dan





Begin forwarded message:

> Date:    Tue, 18 Dec 2007 22:18:48 -0600
> From:    "Gretchen Unger, Ph.D." <[hidden email]>
> Subject: colocalization greater than 100%
>
> Hi all,
>
> I used the division feature of Image J, v. 1.38 to estimate maximum
> colocalization between two optical sections, one was green (reporter
> gene) and the other red (cells). It looked like it worked but in one
> section I calculated a value over 100%. Can anyone explain how that
> might happen?
>
> I transformed final images processed against background into
> grayscale using Photoshop for the Image J calculations. In Image J, I
> calculated a joint (green+red) area by pixel-by-pixel division of
> green/red. This would calculate a maximum colocalization as a pixel
> with any signal above background in both images would divide to a
> "1". Because of drift or jumping or whatever, I had to manually alter
> alignment in some of the joint picture calculations.
>
> Next, I calculated "area fractions" in both the joint and red image
> using the "threshold" feature, applying threshold , and then
> "measuring" following selection of "area fraction" as an attribute.
> Because my final calculation was Colocalization =
> (green/red)*100/red, I selected the thresholding level from my
> denominator using the auto function. I then manually set the
> threshold to the same value (whatever it was) for the area fraction
> measurement of the joint image (the numerator). I got a number around
> 170% (17.25 divided by 10.37),  I have used this method in the past
> and got numbers ranging from 50% to 98%, which seemed reasonable.  I
> don't know why this set of aligned panels exceeds 100% and can't
> explain to anyone how that happened or what it can mean. Should I
> auto-threshold both images ? did I alter a calculation basis somehow
> when I manually aligned the image pair?
>
> Any thoughts or comments would be appreciated.
>
> Gretchen Unger

Dr. Daniel James White BSc. (Hons.) PhD
Senior Microscopist / Image Processing and Analysis
Light Microscopy Facility
Max Planck Institute of Molecular Cell Biology and Genetics
Pfotenhauerstrasse 108
01307 DRESDEN
Germany


New Mobile Number!!!

[phone] (German Mobile)
+49  (0)351 210 2627 (Work phone at MPI-CBG)
+49  (0)351 210 1078 (Fax MPI-CBG LMF)
[phone] (Finnish mobile, only when I'm in Finland)
http://www.bioimagexd.net
http://www.chalkie.org.uk
[hidden email]
( [hidden email] )

[Daniel James White]

> Hi Gretchen,
>
> your method looks interesting,
> but are there are several places where subjectivity creeps in?
>
> To do pixel intensity spatial correlation coloc analysis
> try the coloc plugins from macbiophotonics / imageJ plugins pages
>
> First you must subtract background first - maybe use rolling ball or
> subtract mean of a background area. You have to decise what is  
> background
> and what isn't.
>
> The auto threshold (costes method) in the
> Colocalization Threshold plugin
> will give you a reliable and repeatable non subjective threshold
> bases on Pearsons correlation.
>
> It will then calculate the Manders coefficients (amount of red coloc  
> with green and vice versa
> Manders 1993 i think)
> and show you the 2D histogram or scatter plot,
> which is a good way to visualise the correlation between the 2  
> colour channels.
>
> I would publish the Manders coefficients, the thresholds that are  
> calculated, and the scatter plot (rather than a 2 colour merged  
> image - which is very difficult to interpret scientifically )
>
> Then use the colocalisation test plugin,
> with the Costes method to test the statistical significance of the  
> colo you see.
> The coloc result are not solid without this test , as they could be  
> caused by random
> overlap in a busy image.
>
> Read the coloc docs at macbiophotonics:
> http://www.macbiophotonics.ca/imagej/colour_analysis.htm
>
>
> Other things to consider to get good results.
> 1) you should image beads on the same system to see any spatial and  
> chromatic abberations in the spatial locations of the 2 colour  
> channels over the field of view.
> You might be surprised how bad it is. (seems you are trying to do  
> that already)
>
> 2) on a zeiss confocal there are 3 emsission pinholes to adjust with  
> 100 nm fluorescent beads
> to get the signals coregistered. The engineeers only use the 10x  
> lens and large beads,
> and always leave it in a non optimised state, no good for coloc of  
> small objects.
>
> 3) Olympus and leica confocals only have a single pinhole,
> so als long as you are using a good (expensive) chromatically and  
> field corected lens
> you are in good shape....
>
> 4) ... unless you are using a UV (say 405) laser coming through a  
> different optical fiber
> into the system than the visible laser lines. The collimator must be  
> carefully adjusted to bring the
> "blue" signal illuminated by the lioght from the UV fiber into  
> register with the visible illumination.
>
> 5) 2D images are ok for coloc, but real sampkles are 3D so in order  
> to get a good idea of the whole sample 3D imagging is a good thing.
>
> 6) If you are using a widefield system, consider deconvolution,
> to increase contrast and remove out of focus fuzz,
> but thats a whole kettle of fish in itself)
>
> 7) You might consider an object based coloc method,
> depending on how your biology works.
> Segment out objects, then see how much they overlap
> between channels.
>
> 8) Bleed through of fluorescence from one channel into another is  
> your worst enemy in coloc.
> You can see it clearly in the 2D histogram.
> If you have bleed through, you will likely measure it as  
> colocalisation,
> which is in fact false. Watch out!
>
> 9) BioImageXD is also free like imageJ ,
> and gives the same methods as in the plugins above,
> but in a more coherent user interface.
> If you have problems with that , send me some sample images,
> and we will try to get it to work for you.
>
>
> Say Hi to MN for me, I used to live in Rochester... Mayo Clinic...
>
>
> cheers
>
> Dan
>
>
>
>
>
> Begin forwarded message:
>
>> Date:    Tue, 18 Dec 2007 22:18:48 -0600
>> From:    "Gretchen Unger, Ph.D." <[hidden email]>
>> Subject: colocalization greater than 100%
>>
>> Hi all,
>>
>> I used the division feature of Image J, v. 1.38 to estimate maximum
>> colocalization between two optical sections, one was green (reporter
>> gene) and the other red (cells). It looked like it worked but in one
>> section I calculated a value over 100%. Can anyone explain how that
>> might happen?
>>
>> I transformed final images processed against background into
>> grayscale using Photoshop for the Image J calculations. In Image J, I
>> calculated a joint (green+red) area by pixel-by-pixel division of
>> green/red. This would calculate a maximum colocalization as a pixel
>> with any signal above background in both images would divide to a
>> "1". Because of drift or jumping or whatever, I had to manually alter
>> alignment in some of the joint picture calculations.
>>
>> Next, I calculated "area fractions" in both the joint and red image
>> using the "threshold" feature, applying threshold , and then
>> "measuring" following selection of "area fraction" as an attribute.
>> Because my final calculation was Colocalization =
>> (green/red)*100/red, I selected the thresholding level from my
>> denominator using the auto function. I then manually set the
>> threshold to the same value (whatever it was) for the area fraction
>> measurement of the joint image (the numerator). I got a number around
>> 170% (17.25 divided by 10.37),  I have used this method in the past
>> and got numbers ranging from 50% to 98%, which seemed reasonable.  I
>> don't know why this set of aligned panels exceeds 100% and can't
>> explain to anyone how that happened or what it can mean. Should I
>> auto-threshold both images ? did I alter a calculation basis somehow
>> when I manually aligned the image pair?
>>
>> Any thoughts or comments would be appreciated.
>>
>> Gretchen Unger
>
> Dr. Daniel James White BSc. (Hons.) PhD
> Senior Microscopist / Image Processing and Analysis
> Light Microscopy Facility
> Max Planck Institute of Molecular Cell Biology and Genetics
> Pfotenhauerstrasse 108
> 01307 DRESDEN
> Germany
>
>
> New Mobile Number!!!
>
> [phone] (German Mobile)
> +49  (0)351 210 2627 (Work phone at MPI-CBG)
> +49  (0)351 210 1078 (Fax MPI-CBG LMF)
> [phone] (Finnish mobile, only when I'm in Finland)
> http://www.bioimagexd.net
> http://www.chalkie.org.uk
> [hidden email]
> ( [hidden email] )
>
>
>
>

Dr. Daniel James White BSc. (Hons.) PhD
Senior Microscopist / Image Processing and Analysis
Light Microscopy Facility
Max Planck Institute of Molecular Cell Biology and Genetics
Pfotenhauerstrasse 108
01307 DRESDEN
Germany


New Mobile Number!!!

[phone] (German Mobile)
+49  (0)351 210 2627 (Work phone at MPI-CBG)
+49  (0)351 210 1078 (Fax MPI-CBG LMF)
[phone] (Finnish mobile, only when I'm in Finland)
http://www.bioimagexd.net
http://www.chalkie.org.uk
[hidden email]
( [hidden email] )

[Daniel James White]

Hi Gretchen,

there is no need to saturate the images, just so you can "see"
the lower intensity parts of the image.

Your nice new nikon confocal has nice PMTs that can
digitise intensity over a large range.
There might even be an option to use 12 bit instead of the normal 12.
8 bit range is usually enough (0-255)
Even when the intense parts of the image are not saturated,
the lower intensities will still be recorded reliably over the  
background noise.
You just dont see them as they are nearly back.

The problem here lies in the way your eyes and brain see objects in  
images
that are displayed with a black to green (or red) look up table.
You eyes and brain are evolved to see fruit on trees,
which means bright colours on a dull background.
High contrast.

... that means we are bad at seeing the low intensity parts of the  
image,
using a black to green colour look up table to visualise the scalar  
intensity image data.


Remember the colours are false - all the PMT sees is photons/intensity.
The software lets you choose what colour look up table to visualise the
scalar intensity images with. For Dyes that emit greenish light, you  
usually choose
green, but you could choose anything.

That means there should be an option to use a spectrum or multi colour  
look up table to display the data, so you can then see the low  
intensity data with a bright colour your eyes can see.

When collecting the data at the scope, you should
get the collected image intensities
"Within the Range of the Detector"
Set the PMT voltage/gain and PMT offset using a range indicator / Hi  
Low etc. colour look up table,
so you can see the saturated pixels, and back off the PMT voltage,
and also set the background to close to zero.
This is a must for quantitative imaging,
it is impossible to set PMT voltage/gain and offset properly using a  
normal
black to green look colour look up table.

Note also that you should try to use PMT at about mid voltage,
as they are more noisy at the low and high ends of the possible  
voltage/gain
(electronic noise and amplification noise).
Then change the laser power % to get fairly close to
the signals being within the range of the detector,
then fine tune with PMT voltage and offset.


The signal to noise ratio for confocal imaging at reasonably fast  
frame rates
(around a second or a few seconds)
is limited by photon shot noise.
That means the more photons you collect for each pixel,
the better quality your images will be.

This means to get better quality images,
and more reliable coloc results,
take the images slower!
If you are only taking single optical slices,
then you can easily spend 1 minute
taking a single slice, to get really good quality.

If you are doing 3D imaging, then you probably
want to go faster for each slice,
but you will have more data and so better stats.

Once you have images,
get the bio-formats plugin for imageJ
http://www.loci.wisc.edu/ome/formats.html
which should open your nikon images directly
along with their meta data (pixel size etc).
No need to export as tiff.


to install the bio-formats plugin
download loci-tools.jar
and put it in the imageJ plugins folder
then restart imageJ
Next time you run ImageJ, a new LOCI submenu with several plugins will  
appear in the Plugins menu, including the Bio-Formats Importer and Bio-
Formats Exporter.


Leave photoshop out of the equation. Its not really for science.
You can do everything you need to do in ImageJ,
and you will know exactly what you are doing to your images,
rather than using "a black box".
You will never have to buy it either!
Never use automatic image enhance methods,
they screw up the quantitative nature of your data.
Don't alter brightness and contrast, so similar reasons.
Get this right at the microscope, and you wont need to touch it after.

Now you have the images open in ImageJ,
use the coloc threshold plugin
and look at the 2D scatter plot.
That tells you much about the relationship between the intensity of  
the 2 colour channels over space. Much more informative than a colour  
merge image (which i think is usually misleading and mostly useless)

We are developing a Nikon reader for BioImageXD,
but not sure if it will open your data yet.
We might need to get a small sample data set from you to make it work...
for now you might have to import tiff images exported from
imageJ or the nikon software.


I will have a quick look at your data and get back to you



cheers

Dan







On Dec 20, 2007, at 5:44 AM, Gretchen Unger, Ph.D. wrote:

> Hi Dan,
>
> Thank you for your interest. This is my Christmas  present.
>
> I am using a Nikon C1si spectral confocal at the St. Paul Imaging,  
> but I am working in conventional mode as I have not mastered  
> spectral yet (working on it, though). I know it uses PMTs in  
> spectral mode, so I'll bet it using PMTs in conventional.
>
> I save all my images as bit mapped Tif so I can work with them in  
> photoshop or image J. I definitely did clip my images as I am  
> interested in behavior on the low end. I am a drug delivery  
> scientist using nanoparticles for delivery of plasmids. In these  
> reporter gene studies, I am comparing the green renilla luciferase  
> expression from the plasmid against the dsRed-+ tumor cells. I  
> calculate a background series developed with nonspecific IgG. I am  
> just trying to figure out the simplest thing to do to estimate  
> colocalization and again I am interested in the low end of behavior,  
> any kind of expression corresponding with any kind of expression. As  
> I mentioned, I was wondering if the problem was, that I didn't  
> automatic threshold in both. I'm hesitant to do that.
>
> Here is a sample figure and there is more to do. I have higher  
> background in these pics as we forgot to use gt-fab for mouse-on-
> mouse. Background is much better when we do that  but I haven't shot  
> those yet. I calculated off the unmanipulated (version B) turned  
> into grayscale. I hope this makes some sense.
>
> I will be looking up and studying all of the new terms and programs  
> you have mentioned.
>
> What are doing in Germany? Do you like Germany?
>
> Thanks,
> Gretchen  Unger
>
>
>>> Hi,
>>>
>>> I forgot to mention,
>>>
>>> if the images are intensity saturated then your results using he  
>>> imageJ plugins i described will be wrong.
>>>
>>> if you clip the high intensity data, which usually the most  
>>> important
>>> as you are interested in the brightest things,
>>> then you obviously lose that info. thats a bad thing.
>>>
>>> On confocal systems using PMT detectors
>>> we teach people to use HiLo look up tables where blue is 0 and red  
>>> is 255
>>> with grey scale in between,
>>> as they collect the images at the scope,
>>> so they get the background / offset at close to 0 (speckled blue/
>>> back)
>>> and make sure the highest intensities are not clipped off (no red  
>>> pixels)
>>>
>>> wide field images collected with a ccd camera must always be  
>>> background subtracted.
>>>
>>> let me know of you need help with BioImageXD coloc function.
>>> What is your native image format?
>>> from a confocal? Which?
>>>
>>> you might be interested to read and digest the attached articles.
>>> We use these as the basis for our teaching of pixel intensity  
>>> based coloc,
>>> and these are the methods used in BioImageXD and the imageJ plugins
>>>
>>
>>
>> Attachment converted: Firenze:Costes_etalColoc.pdf (PDF /«IC»)  
>> (00165825)
>> Attachment converted: Firenze:manders.pdf (PDF /«IC») (00165826)
>>>
>>>
>>> cheers
>>>
>>> Dan
>>>
>>>>
>>>>> Hi Gretchen,
>>>>>
>>>>> your method looks interesting,
>>>>> but are there are several places where subjectivity creeps in?
>>>>>
>>>>> To do pixel intensity spatial correlation coloc analysis
>>>>> try the coloc plugins from macbiophotonics / imageJ plugins pages
>>>>>
>>>>> First you must subtract background first - maybe use rolling  
>>>>> ball or
>>>>> subtract mean of a background area. You have to decise what is
>>>>> background
>>>>> and what isn't.
>>>>>
>>>>> The auto threshold (costes method) in the
>>>>> Colocalization Threshold plugin
>>>>> will give you a reliable and repeatable non subjective threshold
>>>>> bases on Pearsons correlation.
>>>>>
>>>>> It will then calculate the Manders coefficients (amount of red  
>>>>> coloc
>>>>> with green and vice versa
>>>>> Manders 1993 i think)
>>>>> and show you the 2D histogram or scatter plot,
>>>>> which is a good way to visualise the correlation between the 2  
>>>>> colour
>>>>> channels.
>>>>>
>>>>> I would publish the Manders coefficients, the thresholds that are
>>>>> calculated, and the scatter plot (rather than a 2 colour merged  
>>>>> image
>>>>> - which is very difficult to interpret scientifically )
>>>>>
>>>>> Then use the colocalisation test plugin,
>>>>> with the Costes method to test the statistical significance of the
>>>>> colo you see.
>>>>> The coloc result are not solid without this test , as they could  
>>>>> be
>>>>> caused by random
>>>>> overlap in a busy image.
>>>>>
>>>>> Read the coloc docs at macbiophotonics:
>>>>> http://www.macbiophotonics.ca/imagej/colour_analysis.htm
>>>>>
>>>>>
>>>>> Other things to consider to get good results.
>>>>> 1) you should image beads on the same system to see any spatial  
>>>>> and
>>>>> chromatic abberations in the spatial locations of the 2 colour
>>>>> channels over the field of view.
>>>>> You might be surprised how bad it is. (seems you are trying to  
>>>>> do that
>>>>> already)
>>>>>
>>>>> 2) on a zeiss confocal there are 3 emsission pinholes to adjust  
>>>>> with
>>>>> 100 nm fluorescent beads
>>>>> to get the signals coregistered. The engineeers only use the 10x  
>>>>> lens
>>>>> and large beads,
>>>>> and always leave it in a non optimised state, no good for coloc of
>>>>> small objects.
>>>>>
>>>>> 3) Olympus and leica confocals only have a single pinhole,
>>>>> so als long as you are using a good (expensive) chromatically and
>>>>> field corected lens
>>>>> you are in good shape....
>>>>>
>>>>> 4) ... unless you are using a UV (say 405) laser coming through a
>>>>> different optical fiber
>>>>> into the system than the visible laser lines. The collimator  
>>>>> must be
>>>>> carefully adjusted to bring the
>>>>> "blue" signal illuminated by the lioght from the UV fiber into
>>>>> register with the visible illumination.
>>>>>
>>>>> 5) 2D images are ok for coloc, but real sampkles are 3D so in  
>>>>> order to
>>>>> get a good idea of the whole sample 3D imagging is a good thing.
>>>>>
>>>>> 6) If you are using a widefield system, consider deconvolution,
>>>>> to increase contrast and remove out of focus fuzz,
>>>>> but thats a whole kettle of fish in itself)
>>>>>
>>>>> 7) You might consider an object based coloc method,
>>>>> depending on how your biology works.
>>>>> Segment out objects, then see how much they overlap
>>>>> between channels.
>>>>>
>>>>> 8) Bleed through of fluorescence from one channel into another  
>>>>> is your
>>>>> worst enemy in coloc.
>>>>> You can see it clearly in the 2D histogram.
>>>>> If you have bleed through, you will likely measure it as  
>>>>> colocalisation,
>>>>> which is in fact false. Watch out!
>>>>>
>>>>> 9) BioImageXD is also free like imageJ ,
>>>>> and gives the same methods as in the plugins above,
>>>>> but in a more coherent user interface.
>>>>> If you have problems with that , send me some sample images,
>>>>> and we will try to get it to work for you.
>>>>>
>>>>>
>>>>> Say Hi to MN for me, I used to live in Rochester... Mayo Clinic...
>>>>>
>>>>>
>>>>> cheers
>>>>>
>>>>> Dan
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Begin forwarded message:
>>>>>
>>>>>> Date:    Tue, 18 Dec 2007 22:18:48 -0600
>>>>>> From:    "Gretchen Unger, Ph.D." <[hidden email]>
>>>>>> Subject: colocalization greater than 100%
>>>>>>
>>>>>> Hi all,
>>>>>>
>>>>>> I used the division feature of Image J, v. 1.38 to estimate  
>>>>>> maximum
>>>>>> colocalization between two optical sections, one was green  
>>>>>> (reporter
>>>>>> gene) and the other red (cells). It looked like it worked but  
>>>>>> in one
>>>>>> section I calculated a value over 100%. Can anyone explain how  
>>>>>> that
>>>>>> might happen?
>>>>>>
>>>>>> I transformed final images processed against background into
>>>>>> grayscale using Photoshop for the Image J calculations. In  
>>>>>> Image J, I
>>>>>> calculated a joint (green+red) area by pixel-by-pixel division of
>>>>>> green/red. This would calculate a maximum colocalization as a  
>>>>>> pixel
>>>>>> with any signal above background in both images would divide to a
>>>>>> "1". Because of drift or jumping or whatever, I had to manually  
>>>>>> alter
>>>>>> alignment in some of the joint picture calculations.
>>>>>>
>>>>>> Next, I calculated "area fractions" in both the joint and red  
>>>>>> image
>>>>>> using the "threshold" feature, applying threshold , and then
>>>>>> "measuring" following selection of "area fraction" as an  
>>>>>> attribute.
>>>>>> Because my final calculation was Colocalization =
>>>>>> (green/red)*100/red, I selected the thresholding level from my
>>>>>> denominator using the auto function. I then manually set the
>>>>>> threshold to the same value (whatever it was) for the area  
>>>>>> fraction
>>>>>> measurement of the joint image (the numerator). I got a number  
>>>>>> around
>>>>>> 170% (17.25 divided by 10.37),  I have used this method in the  
>>>>>> past
>>>>>> and got numbers ranging from 50% to 98%, which seemed  
>>>>>> reasonable.  I
>>>>>> don't know why this set of aligned panels exceeds 100% and can't
>>>>>> explain to anyone how that happened or what it can mean. Should I
>>>>>> auto-threshold both images ? did I alter a calculation basis  
>>>>>> somehow
>>>>>> when I manually aligned the image pair?
>>>>>>
>>>>>> Any thoughts or comments would be appreciated.
>>>>>>
>>>>>> Gretchen Unger
>>>>>
>>>>> Dr. Daniel James White BSc. (Hons.) PhD
>>>>> Senior Microscopist / Image Processing and Analysis
>>>>> Light Microscopy Facility
>>>>> Max Planck Institute of Molecular Cell Biology and Genetics
>>>>> Pfotenhauerstrasse 108
>>>>> 01307 DRESDEN
>>>>> Germany
>>>>>
>>>>>
>>>>> New Mobile Number!!!
>>>>>
>>>>> [phone] (German Mobile)
>>>>> +49  (0)351 210 2627 (Work phone at MPI-CBG)
>>>>> +49  (0)351 210 1078 (Fax MPI-CBG LMF)
>>>>> [phone] (Finnish mobile, only when I'm in Finland)
>>>>> http://www.bioimagexd.net
>>>>> http://www.chalkie.org.uk
>>>>> [hidden email]
>>>>> ( [hidden email] )
>>>>
>>>>
>>>> --
>>>> Joel B. Sheffield, Ph.D.
>>>> Biology Department, Temple University
>>>> 1900 North 12th Street
>>>> Philadelphia, PA 19122
>>>> [hidden email]
>>>> [phone], fax [phone]
>>>> http://astro.temple.edu/~jbs
>>>>
>>>
>>> Dr. Daniel James White BSc. (Hons.) PhD
>>> Senior Microscopist / Image Processing and Analysis
>>> Light Microscopy Facility
>>> Max Planck Institute of Molecular Cell Biology and Genetics
>>> Pfotenhauerstrasse 108
>>> 01307 DRESDEN
>>> Germany
>>>
>>>
>>> New Mobile Number!!!
>>>
>>> [phone] (German Mobile)
>>> +49  (0)351 210 2627 (Work phone at MPI-CBG)
>>> +49  (0)351 210 1078 (Fax MPI-CBG LMF)
>>> [phone] (Finnish mobile, only when I'm in Finland)
>>> http://www.bioimagexd.net
>>> http://www.chalkie.org.uk
>>> [hidden email]
>>> ( [hidden email] )
>>>
>>>
>>>
>>
>> Dr. Daniel James White BSc. (Hons.) PhD
>> Senior Microscopist / Image Processing and Analysis
>> Light Microscopy Facility
>> Max Planck Institute of Molecular Cell Biology and Genetics
>> Pfotenhauerstrasse 108
>> 01307 DRESDEN
>> Germany
>>
>>
>> New Mobile Number!!!
>>
>> [phone] (German Mobile)
>> +49  (0)351 210 2627 (Work phone at MPI-CBG)
>> +49  (0)351 210 1078 (Fax MPI-CBG LMF)
>> [phone] (Finnish mobile, only when I'm in Finland)
>> http://www.bioimagexd.net
>> http://www.chalkie.org.uk
>> [hidden email]
>> ( [hidden email] )
>
> <agrn2.tif><aRed2.tif><Result of agrn2.tif><10ngvsIgG  
> B.psd><10ngvsIgG C.jpg>

Dr. Daniel James White BSc. (Hons.) PhD
Senior Microscopist / Image Processing and Analysis
Light Microscopy Facility
Max Planck Institute of Molecular Cell Biology and Genetics
Pfotenhauerstrasse 108
01307 DRESDEN
Germany


New Mobile Number!!!

[phone] (German Mobile)
+49  (0)351 210 2627 (Work phone at MPI-CBG)
+49  (0)351 210 1078 (Fax MPI-CBG LMF)
[phone] (Finnish mobile, only when I'm in Finland)
http://www.bioimagexd.net
http://www.chalkie.org.uk
[hidden email]
( [hidden email] )