Re: quantification of fluorescent signal per a cell

Posted by Shall, Sydney on Mar 13, 2013; 10:22am
URL: http://imagej.273.s1.nabble.com/quantification-of-fluorescent-signal-per-a-cell-tp5002137p5002142.html

On 13/03/2013 02:22, Junsung LEE wrote:

> I want to quantify fluorescent signal (punctiform, like cellular vesicles) in one cell, but I don't know how. I have tried a few things as follows (but all of them were not successful) :
>
>
> 1. first, I selected one cell with 'polygon selections tool', and tried to analyse fluorescent signal with 'RGB profile (Plugins>Graphics>RGB profile)'. But I failed because 'RGB profile' is working only with 'line or rectangular selections tool'. I can apply 'rectangular selections tool' in some cases, but when cells are overlapped partially, it is not applicable.
>
>> is there any ways to make 'RGB profile' work with 'polygon selections tool' or 'freehand selections tool'?
>
> 2. second, I marked fluorescent signals (they are dot-shaped, as I said above) with 'multi-point selections tool' one by one, and analysed them with 'measure (analyse>measure (ctrl+M)'. This method is working but marking many fluorescent dots one by one is really hard.
>
>> I really wonder how to select fluorescent signals (dots) automatically. Once I designate one dot as signal, it is possible to select dots automatically, I think, but I don't know how.
>
> I will be appreciated if you tell me how. If appropriate method is exist except what I said above, please don't hesitate to   let me know. Please refer to attached image file.
>
> Thanks in advance.
>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

Dear Junsung,
The answer that you need was posted yesterday by myself and someone else.
I copy below the two answers.

Dear Rona,
There is a plug-in for ImageJ crafted by Dr Pawel Znojek, which
essentially aims to do this.
He constructed it for DAPI and a fluorescent  antibody to a DNA repair
protein.
You should be able to adapt this easily for your purposes.
The plugin is available from the following site:
http://www.pzfociez.com/ .
I would recommend that you should add a plugin constructed by Gabriel
Landini, at the University of Birmingham, England, which helps considerably.
This you can find on the ImageJ site.

If you proceed down this path, I should be very grateful to you if you
could let me know how things develop for you.

With best wishes,

Sydney

hi
you can try the color deconvolution way ( landini's plugin)
or the color segmentation,
moreover, is available online a software developed for ki-67 nuclear staining
quantificationhttp://breast-cancer-research.com/content/12/4/R56
maybe at 40 x magnification it could work.
Let me know
Good luck

carlo  bologna Italy


With best wishes,
Sydney

--
Professor Sydney Shall,
Department of Haematological Medicine,
King's College London,
Medical School,
123 Coldharbour Lane,
LONDON SE5 9NU,
Tel & Fax: [phone],
E-Mail: sydney.shall,
[correspondents outside the College should add; @kcl.ac.uk]
www.kcl.ac.uk

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