Measuring colocalisation in confocal microscopy images

[Seneviratne, Anusha]

Hi everyone,

In the attached image I would like to be able to measure any areas stained red as a percentage of the cells stained green, i.e. where they colocalise. How can I do this using ImageJ? Also I have tried to install the colocalisation plugins into ImageJ on my Mac and it doesn't work. I was copying the .class files into the plugins folder and then told the computer to always open the files using ImageJ. Is there anything else I need to do to make the plugins appear when I open ImageJ?

I would be grateful for any help.

Thanks,
Anusha

[cid:AB2171EF-2ED2-4E8F-809E-A535CD3388B7]

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Robert Baer]

If you have not read it, one place to start is with Tony Collins MBF
ImageJ mannual: http://www.macbiophotonics.ca/imagej/colour_analysis.htm

Another useful place to look is:
http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:jacop_2.0:just_another_colocalization_plugin:start
I just downloaded JACoP and dropped it in my FIJI plugins folder,
restarted FIJI, and there it was working fine.  You don't say exactly
which plugin you tried so I can't vouch for other specific plugins.

Hope this helps,

Rob

On 9/9/2012 8:42 AM, Seneviratne, Anusha wrote:

> Hi everyone,
>
> In the attached image I would like to be able to measure any areas stained red as a percentage of the cells stained green, i.e. where they colocalise. How can I do this using ImageJ? Also I have tried to install the colocalisation plugins into ImageJ on my Mac and it doesn't work. I was copying the .class files into the plugins folder and then told the computer to always open the files using ImageJ. Is there anything else I need to do to make the plugins appear when I open ImageJ?
>
> I would be grateful for any help.
>
> Thanks,
> Anusha
>
> [cid:AB2171EF-2ED2-4E8F-809E-A535CD3388B7]
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Tom Kazimiers]

Hi Anusha, hi Robert,

On 09.09.2012 22:52, Robert Baer wrote:
> If you have not read it, one place to start is with Tony Collins MBF
> ImageJ mannual: -

Thanks for the link. It gives a good overview on some of the available
plugins (not all, though).

> On 9/9/2012 8:42 AM, Seneviratne, Anusha wrote:
>> In the attached image I would like to be able to measure any areas
>> stained red as a percentage of the cells stained green, i.e. where
>> they colocalise. How can I do this using ImageJ?

As an alternative to what you have tried, you could also have a look at
Fiji (Fiji Is Just ImageJ -- batteries included). Part of this ImageJ
distribution is the plugin colocalization analysis plugin Coloc 2 [1].

It supports Li historgrams and Li's ICQ value, Spearman's rank
correlation, Mander's correlation, Pearson's correlation, displaying a
2d intensity diagram (scatter plot) and Costes' significance test.
Auto-thresholding is used for all those algorithms. Also, you can use
ROIs and masks (b/w images with same dimensionality to mark areas to
analyse → kind of 3D ROIs).

>> Also I have tried to
>> install the colocalisation plugins into ImageJ on my Mac and it
>> doesn't work. I was copying the .class files into the plugins folder
>> and then told the computer to always open the files using ImageJ. Is
>> there anything else I need to do to make the plugins appear when I
>> open ImageJ?

Like Robert asked: What plugins were you actually testing?

As far es I know of, opening .class files with ImageJ won't load the
plugin. Therefore, you should not need to tell your operating system to
open .class files with ImageJ. It should be enough to place the .jar or
.class file in the plugins folder. Also the plugin files (say
JaCoP_.jar) need to have an underscore in its name to be picked up as a
plugin automatically. They also need to start with a capital letter. Of
course, you can also install the plugins manually (Plugins menu >
Install plugin). (see [2])

Cheers,
Tom

[1] http://fiji.sc/wiki/index.php/Colocalization_Analysis
[2] http://rsbweb.nih.gov/ij/docs/guide/146-16.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Seneviratne, Anusha]

Hi Robert and Tom,

Thanks a lot to both of you for your advice. I was trying to install the bundle of colocalisation plugins on the Wright Cell Imaging Facility webpage found here http://www.uhnresearch.ca/facilities/wcif/fdownload.html 

I followed the Mac instructions for installing the plugins. Maybe I am blind but I looked in all the menus when opening ImageJ and I could not see where to select any of the colocalisation plugins. I added them into the ImageJ plugins folder (Finder>Library>Application Support>ImageJ>plugins). As far as I can see all the .class file names start with a capital letter and they have an underscore. In ImageJ I also tried Plugins>Install plugin as you say Tom and it says "No plugins found".

Alternatively I will try FIJI and see whether it works better for me.

Thanks again for your help!!

Anusha



-----Original Message-----
From: Tom Kazimiers [mailto:[hidden email]]
Sent: 10 September 2012 10:17
To: [hidden email]
Cc: Robert Baer; Seneviratne, Anusha
Subject: Re: Measuring colocalisation in confocal microscopy images

Hi Anusha, hi Robert,

On 09.09.2012 22:52, Robert Baer wrote:
> If you have not read it, one place to start is with Tony Collins MBF
> ImageJ mannual: -

Thanks for the link. It gives a good overview on some of the available plugins (not all, though).

> On 9/9/2012 8:42 AM, Seneviratne, Anusha wrote:
>> In the attached image I would like to be able to measure any areas
>> stained red as a percentage of the cells stained green, i.e. where
>> they colocalise. How can I do this using ImageJ?

As an alternative to what you have tried, you could also have a look at Fiji (Fiji Is Just ImageJ -- batteries included). Part of this ImageJ distribution is the plugin colocalization analysis plugin Coloc 2 [1].

It supports Li historgrams and Li's ICQ value, Spearman's rank correlation, Mander's correlation, Pearson's correlation, displaying a 2d intensity diagram (scatter plot) and Costes' significance test.
Auto-thresholding is used for all those algorithms. Also, you can use ROIs and masks (b/w images with same dimensionality to mark areas to analyse → kind of 3D ROIs).

>> Also I have tried to
>> install the colocalisation plugins into ImageJ on my Mac and it
>> doesn't work. I was copying the .class files into the plugins folder
>> and then told the computer to always open the files using ImageJ. Is
>> there anything else I need to do to make the plugins appear when I
>> open ImageJ?

Like Robert asked: What plugins were you actually testing?

As far es I know of, opening .class files with ImageJ won't load the plugin. Therefore, you should not need to tell your operating system to open .class files with ImageJ. It should be enough to place the .jar or .class file in the plugins folder. Also the plugin files (say
JaCoP_.jar) need to have an underscore in its name to be picked up as a plugin automatically. They also need to start with a capital letter. Of course, you can also install the plugins manually (Plugins menu > Install plugin). (see [2])

Cheers,
Tom

[1] http://fiji.sc/wiki/index.php/Colocalization_Analysis
[2] http://rsbweb.nih.gov/ij/docs/guide/146-16.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html