Comparing image intensities in a volume

[Manuella Clark-Cotton]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hello,

I have a question that I haven't seen addressed in the archives. Apologies
if I've overlooked it.

I am labeling the kinetochore proteins of cultured cells, and I'd like to
be able to compare the intensity of kinetochore fluorescence (in the
nucleus) in the entire volume of two daughter cell pairs. I can acquire a
set of slices from the microscope, but I can't figure out how to measure
intensity in a volume without measuring separately for each slice.
Suggestions on how I would do this using ImageJ/MRI would be greatly
appreciated!

Regards,
Rossie Clark-Cotton

[mmodel]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hello Rossie,

Wide field collects pretty much the entire fluorescence, even from defocused fields. You can check it my measuring the cumulative signal from a bead at different positions of the objective. Only make sure to include defocused areas in the measurement.

Mike Model

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Manuella Clark-Cotton
Sent: Friday, April 26, 2013 10:18 AM
To: [hidden email]
Subject: Comparing image intensities in a volume

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hello,

I have a question that I haven't seen addressed in the archives. Apologies if I've overlooked it.

I am labeling the kinetochore proteins of cultured cells, and I'd like to be able to compare the intensity of kinetochore fluorescence (in the
nucleus) in the entire volume of two daughter cell pairs. I can acquire a set of slices from the microscope, but I can't figure out how to measure intensity in a volume without measuring separately for each slice.
Suggestions on how I would do this using ImageJ/MRI would be greatly appreciated!

Regards,
Rossie Clark-Cotton

[Julio Vazquez]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi Manuela,

If your data is relatively clean (good signal for kinetochores, low background, no signal outside nucleus, you can load the stack of images into imageJ using the import image sequence tool, then threshold the intensity to select kinetochores, and use the Analyze particles tool. This gives you the option to analyze the entire series, and all the data is sent to a text file. You can bring the numbers into Excel and just add up all the relevant values.

If there is background outside the nucleus and/or need to restrict the analysis exclusively to the nucleus, you could threshold the nuclear stack and convert the selection to a mask. You would then need to use the masks to restrict analysis of kinetochores. You can do that by going to Analyze > Set measurements, and redirect the analysis to the kinetochore stack. This will use the nuclear stack as a mask to analyze the kinetochore stack.

ImageJ will not connect data between sections to analyze the 3-D objects. You are still adding up data from each section t get global results, but will be much faster than analysing each section manually.

Apparently, there are attemps at making ImageJ analyze in 3-D, but haven't tried it out. You can check it out here:

http://rsbweb.nih.gov/ij/docs/guide/146-12.html

Hope this helps.


--
Julio Vazquez
Fred Hutchinson Cancer Research Center
Seattle, WA 98109-1024
 
http://www.fhcrc.org

==


On Apr 26, 2013, at 7:17 AM, Manuella Clark-Cotton wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hello,
>
> I have a question that I haven't seen addressed in the archives. Apologies
> if I've overlooked it.
>
> I am labeling the kinetochore proteins of cultured cells, and I'd like to
> be able to compare the intensity of kinetochore fluorescence (in the
> nucleus) in the entire volume of two daughter cell pairs. I can acquire a
> set of slices from the microscope, but I can't figure out how to measure
> intensity in a volume without measuring separately for each slice.
> Suggestions on how I would do this using ImageJ/MRI would be greatly
> appreciated!
>
> Regards,
> Rossie Clark-Cotton

[Julio Vazquez]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi Manuella,

Yes, obviously if you have access to a 3D image analysis package such as Volocity, Imaris, etc, you can do everything in 3D, i.e. obtain the number of kinetochores per nucleus, and get the intensity and volume of each kinetochore. I am guessing that if you asked specifically about ImageJ, it is because you do not have such software... I am not sure what you mean by MRI though...

The main difference with the ImageJ Analyze Particles method I suggested is that ImageJ will give you an aggregate value (total intensity/Volume of all kinetochores), but not a kinetochore count or individual kinetochore measurements, since the same particle will be present in several images. However, if the data is clean, you could do a "sum" projection and analyze the projection using the same method, as long as you don't get overlap of different kinetochores. In this case, you would also get a count and individul measurements. If you have signal outside of the nucleus, you would need to mask the region outside the nucleus for each section first (i.e. convert to black) before doing the projection. You could do that by thresholding the nucleus, converting to a selection (Edit>Selection>Create selection), transfering the selection to your kinetochore images (use teh ROI manager for that), and clearing outside.

There is actually a plugin in ImageJ that measures 3D objects: Object Counter 3D. However, I don't know if you can use the nucleus image stack as a mask.... if you have signal outside the nucleus that you need to exclude, you probably may need to clear the regions outside the nucleus as above, and then run the 3D object counter plugin.

Hope all this makes sense.

--
Julio Vazquez
Fred Hutchinson Cancer Research Center
Seattle, WA 98109-1024

http://www.fhcrc.org/


On Apr 26, 2013, at 7:17 AM, Manuella Clark-Cotton wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hello,
>
> I have a question that I haven't seen addressed in the archives. Apologies
> if I've overlooked it.
>
> I am labeling the kinetochore proteins of cultured cells, and I'd like to
> be able to compare the intensity of kinetochore fluorescence (in the
> nucleus) in the entire volume of two daughter cell pairs. I can acquire a
> set of slices from the microscope, but I can't figure out how to measure
> intensity in a volume without measuring separately for each slice.
> Suggestions on how I would do this using ImageJ/MRI would be greatly
> appreciated!
>
> Regards,
> Rossie Clark-Cotton

[Manuella Clark-Cotton]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Thanks for all of the suggestions. I mentioned MRI (Montpellier RIO Imaging),
which I read (somewhere, don't recall where) would be useful, but it sounds as if
I can do this with ImageJ. Appreciate the help!