ImageJ

fluorescence qunatification comparison

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

5 messages Options

Judy Trogadis-2

fluorescence qunatification comparison

Hi All

I am comparing surface receptor staining of 2 sets of fluorecently stained samples. The first is a control sample, the second set has been treated and we are looking at potential clustering of the receptors following treatment.

How can I quantitate this? If clustered, simple thresholding will just show a smaller number of stained pixels, it will not indicate a shift to fewer but brighter pixels. Or can it?

Thank you,

Judy

Judy Trogadis
Bio-Imaging Coordinator
Keenan Research Centre, St. Michael's
209 Victoria Street
Toronto M5B 1T8, Canada
office: 416-864-6060 ext. 77612
imaging facility: ext. 77434
cell: 416-254-9330
[hidden email]<mailto:[hidden email]>

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

John Oreopoulos

Re: fluorescence qunatification comparison

Hi Judy, below I copy and paste my answer to this common type of question that shows up every now and then on the ImageJ listserver:

Your question is a very general one that gets asked about 2-4 times a year on the ImageJ email listserver. (There really should be a permanent link on the ImageJ website FAQ for this question.) Quantifying and comparing fluorescence intensities acquired with a microscope is possible using ImageJ, but you have to make sure the data is collected carefully and properly first (and this is usually not trivial!). I would highly recommend you get your hands on and read the following published articles on this topic:

Pawley, J., The 39 steps: A cautionary tale of quantitative 3-D fluorescence microscopy. Biotechniques, 2000. 28(5): p. 884-888.

North, A.J., Seeing is believing? A beginners' guide to practical pitfalls in image acquisition. Journal of Cell Biology, 2006. 172(1): p. 9-18.

Waters, J.C., Interpreting fluorescence microscopy images and measurements. Cell, 2008.

Waters, J.C., Accuracy and precision in quantitative fluorescence microscopy. Journal of Cell Biology, 2009. 185(7): p. 1135-1148.

Then, after that, take a look at the excellent tutorials on the ImageJ Wiki and the FIJI website (and others as well):

http://imagejdocu.tudor.lu/doku.php?id=video:start
http://pacific.mpi-cbg.de/wiki/index.php/Colocalization#Colocalization_Analysis_-_What_is_Colocalization_anyway.3F
http://rsbweb.nih.gov/ij/docs/examples/index.html
http://www.fmhs.auckland.ac.nz/sms/biru/_docs/Image_Analysis_Basics.pdf

There was another excellent link someone posted a while back about basic intensity measurements via ImageJ, but now I can't find it. Try searching the email archive and you might find it.

Unfortunately, there is no one recipe for analyzing the data that you collect with your microscope system since every imaging situation and sample is different and usually the researcher has very specific questions they want to answer with the analysis. You'll have to find out what methods and plugins work best for your data. Search the plugin list and see if someone has developed a specific tool for your application already.

Cheers, and good luck!

And to my normal message, I would add that in your case, it sounds like a specialized method such as Image Correlation Spectroscopy (ICS) or Number and Brightness analysis (N&B) might work for you. Unfortunately, neither of these analysis methods to the best of my knowledge have been implemented in ImageJ yet.

John Oreopoulos
Staff Scientist
Spectral Applied Research Inc.
A Division of Andor Technology
Richmond Hill, Ontario
Canada
www.spectral.ca

On 2014-01-26, at 1:02 PM, Judy Trogadis wrote:

> Hi All
>
> I am comparing surface receptor staining of 2 sets of fluorecently stained samples. The first is a control sample, the second set has been treated and we are looking at potential clustering of the receptors following treatment.
>
> How can I quantitate this? If clustered, simple thresholding will just show a smaller number of stained pixels, it will not indicate a shift to fewer but brighter pixels. Or can it?
>
> Thank you,
>
> Judy
>
>
> Judy Trogadis
> Bio-Imaging Coordinator
> Keenan Research Centre, St. Michael's
> 209 Victoria Street
> Toronto M5B 1T8, Canada
> office: 416-864-6060 ext. 77612
> imaging facility: ext. 77434
> cell: 416-254-9330
> [hidden email]<mailto:[hidden email]>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

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

Judy Trogadis-2

Re: fluorescence qunatification comparison

Hi John

Thank you for your thoughtful and detailed response. I understand the issue and do my best to capture images of which Jim Pawley would be proud.

What I was hoping to see is a plugin in ImageJ that allows a user to sub-divide the analysis results into bins of increasing intensity which can then be plotted - bin number against intensity and the slope of the graph compared between the control and treated samples. I did this in a paper years ago with some interesting results - alas, it was with in-house software which no longer exists. The N&B analysis sounds interesting.

Thanks,
Judy

________________________________________
From: ImageJ Interest Group [[hidden email]] on behalf of John Oreopoulos [[hidden email]]
Sent: 26 January 2014 16:13
To: [hidden email]
Subject: Re: fluorescence qunatification comparison

Hi Judy, below I copy and paste my answer to this common type of question that shows up every now and then on the ImageJ listserver:

Your question is a very general one that gets asked about 2-4 times a year on the ImageJ email listserver. (There really should be a permanent link on the ImageJ website FAQ for this question.) Quantifying and comparing fluorescence intensities acquired with a microscope is possible using ImageJ, but you have to make sure the data is collected carefully and properly first (and this is usually not trivial!). I would highly recommend you get your hands on and read the following published articles on this topic:

Pawley, J., The 39 steps: A cautionary tale of quantitative 3-D fluorescence microscopy. Biotechniques, 2000. 28(5): p. 884-888.

North, A.J., Seeing is believing? A beginners' guide to practical pitfalls in image acquisition. Journal of Cell Biology, 2006. 172(1): p. 9-18.

Waters, J.C., Interpreting fluorescence microscopy images and measurements. Cell, 2008.

Waters, J.C., Accuracy and precision in quantitative fluorescence microscopy. Journal of Cell Biology, 2009. 185(7): p. 1135-1148.

Then, after that, take a look at the excellent tutorials on the ImageJ Wiki and the FIJI website (and others as well):

http://imagejdocu.tudor.lu/doku.php?id=video:start
http://pacific.mpi-cbg.de/wiki/index.php/Colocalization#Colocalization_Analysis_-_What_is_Colocalization_anyway.3F
http://rsbweb.nih.gov/ij/docs/examples/index.html
http://www.fmhs.auckland.ac.nz/sms/biru/_docs/Image_Analysis_Basics.pdf

There was another excellent link someone posted a while back about basic intensity measurements via ImageJ, but now I can't find it. Try searching the email archive and you might find it.

Unfortunately, there is no one recipe for analyzing the data that you collect with your microscope system since every imaging situation and sample is different and usually the researcher has very specific questions they want to answer with the analysis. You'll have to find out what methods and plugins work best for your data. Search the plugin list and see if someone has developed a specific tool for your application already.

Cheers, and good luck!

And to my normal message, I would add that in your case, it sounds like a specialized method such as Image Correlation Spectroscopy (ICS) or Number and Brightness analysis (N&B) might work for you. Unfortunately, neither of these analysis methods to the best of my knowledge have been implemented in ImageJ yet.

John Oreopoulos
Staff Scientist
Spectral Applied Research Inc.
A Division of Andor Technology
Richmond Hill, Ontario
Canada
www.spectral.ca

On 2014-01-26, at 1:02 PM, Judy Trogadis wrote:

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

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

John Oreopoulos

Re: fluorescence qunatification comparison

Hi Judy,

Now it's a bit clearer what you're trying to do. Take a look at these ImageJ functions:

http://rsbweb.nih.gov/ij/docs/guide/146-30.html#sub:Distribution...
http://rsbweb.nih.gov/ij/docs/guide/146-30.html#sub:Histogram

Cheers,

John Oreopoulos
Staff Scientist
Spectral Applied Research Inc.
A Division of Andor Technology
Richmond Hill, Ontario
Canada
www.spectral.ca

On 2014-01-26, at 6:16 PM, Judy Trogadis wrote:

> Hi John
>
> Thank you for your thoughtful and detailed response. I understand the issue and do my best to capture images of which Jim Pawley would be proud.
>
> What I was hoping to see is a plugin in ImageJ that allows a user to sub-divide the analysis results into bins of increasing intensity which can then be plotted - bin number against intensity and the slope of the graph compared between the control and treated samples. I did this in a paper years ago with some interesting results - alas, it was with in-house software which no longer exists. The N&B analysis sounds interesting.
>
> Thanks,
> Judy
>
> ________________________________________
> From: ImageJ Interest Group [[hidden email]] on behalf of John Oreopoulos [[hidden email]]
> Sent: 26 January 2014 16:13
> To: [hidden email]
> Subject: Re: fluorescence qunatification comparison
>
> Hi Judy, below I copy and paste my answer to this common type of question that shows up every now and then on the ImageJ listserver:
>
> Your question is a very general one that gets asked about 2-4 times a year on the ImageJ email listserver. (There really should be a permanent link on the ImageJ website FAQ for this question.) Quantifying and comparing fluorescence intensities acquired with a microscope is possible using ImageJ, but you have to make sure the data is collected carefully and properly first (and this is usually not trivial!). I would highly recommend you get your hands on and read the following published articles on this topic:
>
> Pawley, J., The 39 steps: A cautionary tale of quantitative 3-D fluorescence microscopy. Biotechniques, 2000. 28(5): p. 884-888.
>
> North, A.J., Seeing is believing? A beginners' guide to practical pitfalls in image acquisition. Journal of Cell Biology, 2006. 172(1): p. 9-18.
>
> Waters, J.C., Interpreting fluorescence microscopy images and measurements. Cell, 2008.
>
> Waters, J.C., Accuracy and precision in quantitative fluorescence microscopy. Journal of Cell Biology, 2009. 185(7): p. 1135-1148.
>
> Then, after that, take a look at the excellent tutorials on the ImageJ Wiki and the FIJI website (and others as well):
>
> http://imagejdocu.tudor.lu/doku.php?id=video:start
> http://pacific.mpi-cbg.de/wiki/index.php/Colocalization#Colocalization_Analysis_-_What_is_Colocalization_anyway.3F
> http://rsbweb.nih.gov/ij/docs/examples/index.html
> http://www.fmhs.auckland.ac.nz/sms/biru/_docs/Image_Analysis_Basics.pdf
>
> There was another excellent link someone posted a while back about basic intensity measurements via ImageJ, but now I can't find it. Try searching the email archive and you might find it.
>
> Unfortunately, there is no one recipe for analyzing the data that you collect with your microscope system since every imaging situation and sample is different and usually the researcher has very specific questions they want to answer with the analysis. You'll have to find out what methods and plugins work best for your data. Search the plugin list and see if someone has developed a specific tool for your application already.
>
> Cheers, and good luck!
>
> And to my normal message, I would add that in your case, it sounds like a specialized method such as Image Correlation Spectroscopy (ICS) or Number and Brightness analysis (N&B) might work for you. Unfortunately, neither of these analysis methods to the best of my knowledge have been implemented in ImageJ yet.
>
> John Oreopoulos
> Staff Scientist
> Spectral Applied Research Inc.
> A Division of Andor Technology
> Richmond Hill, Ontario
> Canada
> www.spectral.ca
>
>
> On 2014-01-26, at 1:02 PM, Judy Trogadis wrote:
>
>> Hi All
>>
>> I am comparing surface receptor staining of 2 sets of fluorecently stained samples. The first is a control sample, the second set has been treated and we are looking at potential clustering of the receptors following treatment.
>>
>> How can I quantitate this? If clustered, simple thresholding will just show a smaller number of stained pixels, it will not indicate a shift to fewer but brighter pixels. Or can it?
>>
>> Thank you,
>>
>> Judy
>>
>>
>> Judy Trogadis
>> Bio-Imaging Coordinator
>> Keenan Research Centre, St. Michael's
>> 209 Victoria Street
>> Toronto M5B 1T8, Canada
>> office: 416-864-6060 ext. 77612
>> imaging facility: ext. 77434
>> cell: 416-254-9330
>> [hidden email]<mailto:[hidden email]>
>>
>> --
>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
>
>
>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

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

Glen MacDonald-2

Re: fluorescence qunatification comparison

In reply to this post by Judy Trogadis-2

Judy,
the Integrated Intensity measurement may be more useful than a mean. It sums intensities, which may be more useful for addressing the issue of small bright region vs. similar amount of label dispersed over a larger volume. But always there are the caveats in the references that John provided.

Glen MacDonald
Core for Communication Research
Virginia Merrill Bloedel Hearing Research Center
Cellular Morphology Core
Center on Human Development and Disability
Box 357923
University of Washington
Seattle, WA 98195-7923 USA
(206) 616-4156
[hidden email]

On Jan 26, 2014, at 3:16 PM, Judy Trogadis <[hidden email]> wrote:

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