Normalizing background and fluorescence intensity

[Dawn Koh]

Hi all,

I'm currently working on labeling DNA with Alexa Fluor Dye 488. In my images
I can see nuclei
with maximum fluorescence at S-phase and slight fluorescence when undergoing
DNA repair.
This is actually another method to assay for unscheduled DNA synthesis (UDS)
instead of using radioisotopes.

My problem begins when I try to compare images from different views or cover
slips.
As I try to compare nuclei at S phase from different cover slips,
I notice that the fluorescence intensity slightly stronger for one cover
slip and weaker for the other.
I need both cover slips to have the same intensity in order to compare the
nuclei undergoing DNA repair.
Does anyone know how I can normalize the images on Image J so that they can
be compared fairly?

Thanks!

Best Regards
Dawn

[Mario Faretta]

Dear Dawn,
there are a lot of factors to be considered in your problem. Variations in
intensity can be related to the way you acquire images (depending on the
setting of your microscope, filters, lamp alignment...) or to the preparation
of the samples (efficiency of the staining, batch of reagents,...).
It could be a tricky stuff to compare "quantitatively" different coverslips
and preparations.
To correct the first factors you can try to apply a "flatfield" correction to
make illumination homogeneous (in the plugin section on the ImageJ website you
can find some solutions with good explanations and you can also search archive
of the list for the topic). You can use empty fields of view (no cells) to
acquire images for background subtraction using the image calculator or to
apply one of the background functions.
For the second story, i.e. normalize signals for comparative measurements, the
best thing is probably to have an internal "biological" standard to refer as
unit of measure of your experiments...
Hope it helps
Mario

 Dawn Koh ([hidden email]) wrote:
 >
 > Hi all,
 >
 > Im currently working on labeling DNA with Alexa Fluor Dye 488. In
my images
 > I can see nuclei
 > with maximum fluorescence at S-phase and slight fluorescence when
undergoing
 > DNA repair.
 > This is actually another method to assay for unscheduled DNA
synthesis (UDS)
 > instead of using radioisotopes.
 >
 > My problem begins when I try to compare images from different
views or cover
 > slips.
 > As I try to compare nuclei at S phase from different cover slips,
 > I notice that the fluorescence intensity slightly stronger for one
cover
 > slip and weaker for the other.
 > I need both cover slips to have the same intensity in order to
compare the
 > nuclei undergoing DNA repair.
 > Does anyone know how I can normalize the images on Image J so that
they can
 > be compared fairly?
 >
 > Thanks!
 >
 > Best Regards
 > Dawn
 >

--
Mario Faretta
Department of Experimental Oncology
European Institute of Oncology
c/o IFOM-IEO Campus for Oncogenomics
via Adamello 16
20139 Milan
Italy
Phone: ++39-02574303054
email: [hidden email]
http://www.ifom-ieo-campus.it



[- Il Tuo 5 per Mille a favore della Ricerca dell'Istituto Europeo di
Oncologia

Tutti coloro che presentano il modello Unico, il modello 730 o più
semplicemente che ricevono dal proprio datore di lavoro il modello CUD, hanno
la facoltà di scegliere la destinazione del proprio 5 per mille.

Nella casella riservata al Finanziamento agli Enti della Ricerca Sanitaria
inserisci il codice fiscale dello IEO (08 69 14 40 153) ed apponi la Tua
firma.
Il Tuo 5 per Mille verrà destinato alla ricerca contro il cancro dell'Istituto
Europeo di Oncologia

NON COSTA NULLA E NON COMPORTA ALCUN AUMENTO DELLE IMPOSTE DA VERSARE
NON È UN'ALTERNATIVA ALL'8 PER MILLE
E' UN GESTO CONCRETO E DI GRANDE VALORE

Per saperne di più vai al sito dello IEO www.ieo.it < http://www.ieo.it >
oppure scrivi a [hidden email] < mailto:[hidden email]>

Segnala ad un amico questa opportunità

 -]

[Janne Hyötylä]

Dear Dawn,

I second what Mario said. But it might be difficult to find an internal  
biological standard for the staining with Alexa488. Is there a possibility  
to stain something in the cytoplasm which has always the same  
accessibility to Alexa488-antibodies (or are you using some other method  
than antibodies?), which does not change during the cell cycle and which  
doesn't interfere with your real signal?

You could possibly use the S-phase nuclei as the intensity standard if the  
following are true:
* You have at least one S-phase nucleus in each image
* You are sure that the S-phase nuclei should always have the same  
intrinsic brightness. I.e. the epitopes have always the same  
accessibility, the number of epitopes is always the same, etc.
* Any brightness change is the same for the whole image, i.e. this would  
be the case if the mounting procedure has an effect on the brightness or  
if the dye itself differs from batch to batch. If for example the  
brightness of the dye somehow depends on the local environment, then this  
is not the case.

 From the above points actually follows that all S-phase nuclei in one  
image should have exactly the same brightness in image after background  
subtraction.

As you can see, to do something quantitative is indeed quite tricky and  
you have to consider all pitfalls carefully, and be aware of the  
assumptions and implications when using any quantitative method.


As for how to do it, once you have your internal brightness standard (e.g.  
the S-phase nuclei), you should do a background subtraction as described  
by Mario (again, many possible ways with their own implications). Then  
measure the intensity of your standard (normally using the ROI manager and  
"Measure..." is good, depends how your images look like) and divide the  
image by this value (Process > Math...). You can convert your image to  
32-bit before normalizing to avoid loss of information.

Best,

Janne



On Wed, 24 Mar 2010 06:34:22 +0100, Mario Faretta  
<[hidden email]> wrote:

> Dear Dawn,
> there are a lot of factors to be considered in your problem. Variations  
> in
> intensity can be related to the way you acquire images (depending on the
> setting of your microscope, filters, lamp alignment...) or to the  
> preparation
> of the samples (efficiency of the staining, batch of reagents,...).
> It could be a tricky stuff to compare "quantitatively" different  
> coverslips
> and preparations.
> To correct the first factors you can try to apply a "flatfield"  
> correction to
> make illumination homogeneous (in the plugin section on the ImageJ  
> website you
> can find some solutions with good explanations and you can also search  
> archive
> of the list for the topic). You can use empty fields of view (no cells)  
> to
> acquire images for background subtraction using the image calculator or  
> to
> apply one of the background functions.
> For the second story, i.e. normalize signals for comparative  
> measurements, the
> best thing is probably to have an internal "biological" standard to  
> refer as
> unit of measure of your experiments...
> Hope it helps
> Mario
>
>  Dawn Koh ([hidden email]) wrote:
>  >
>  > Hi all,
>  >
>  > Im currently working on labeling DNA with Alexa Fluor Dye 488. In
> my images
>  > I can see nuclei
>  > with maximum fluorescence at S-phase and slight fluorescence when
> undergoing
>  > DNA repair.
>  > This is actually another method to assay for unscheduled DNA
> synthesis (UDS)
>  > instead of using radioisotopes.
>  >
>  > My problem begins when I try to compare images from different
> views or cover
>  > slips.
>  > As I try to compare nuclei at S phase from different cover slips,
>  > I notice that the fluorescence intensity slightly stronger for one
> cover
>  > slip and weaker for the other.
>  > I need both cover slips to have the same intensity in order to
> compare the
>  > nuclei undergoing DNA repair.
>  > Does anyone know how I can normalize the images on Image J so that
> they can
>  > be compared fairly?
>  >
>  > Thanks!
>  >
>  > Best Regards
>  > Dawn
>  >