Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

There are a couple of issues to keep in mind:

Spatial concerns:  The more labels you have in your cells, the more
likely they will overlap within the cell, giving a "jumbled" and less
informative appearance (and detract from the beauty).

Spectral concerns:  The more fluorescent labels you have, the more
likely you will encounter spectral bleedthrough issues.  

Coloration of far red dyes:  As for coloration of four-color images, we
at Molecular Probes tend to color the longer-wavelength dyes from
magenta (at 647nm, like Alexa Fluor 647 and Cy5) to deep purple (at
around 700nm, Alexa Fluor 680, for instance).  Some people make the far
red channel white (though this can be confused with points where all
colors overlap) or yellow (a less-used channel for most fluorescent
applications).

Recombining monochrome images:  The method of recombining will differ
from program to program, though I tend to use a combination of MetaMorph
(for initial image capture and simple multi-color overlays) and
PhotoShop (for more complex overlays).

Ethics:  The main concern is to realistically depict what you see by eye
through the microscope.  To that end, manipulate the image only as much
as necessary to match what you saw by eye.  This doesn't apply to many
far red dyes, of course, since you cannot see them by eye.  In that
case, what you "see" by camera is your only guide.  Any manipulation of
that image, then, is suspect, and should be done very judiciously.

RGB vs CMYK:  Most of my work is in RGB.  Conversion to CMYK can "muddy"
the image, and take some skill to do well (blue channels, for instance,
gets "muddy", in which case you have to convert to cyan).  I've only had
to do conversions when the images go for publications (as printers use
CMYK).


Jason A. Kilgore
Cell Biology
Molecular Probes / Invitrogen
Eugene, OR
 
"You must be the change you wish to see in the world."
              -- Mahatma Ghandi


Uriel TK wrote:
> Search the CONFOCAL archive at
> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
> Friends:
>
> Hi!
> We are mainly a flow cytometry lab and lately we are using more and 
> more (and will indeed continue increasing) our imaging cytometry work.

> Thus I've got to start getting serious about my images! The 
> preparation of slides is the small problem so far, I've recently 
> performed 4 color (DAPI, FITC/A488, rhod, DiD/Cy5) + transmitted light

> preparations that looked great... each one on its own. My main 
> headache now working with the individual images and getting a final 
> result. before I dig into it I wanted to ask for some couple of 
> directions:
>
> 1) Is there any "standard" for 4 colors? RGB obviously uses only 3. is

> it CYMK?
> 2) What is ethically/scientifically accepted as far as manipulating 
> the images? Given enough time and imagination (or rather, 
> obstination!) tweaking thresholds, contrasts, intensity and the like 
> you can make a lot of images look like "something else" or "something 
> better".
> 3) I guess that I should pick one program and become adept at it. For 
> advanced but not cutting edge work, what do you say: ImageJ or 
> Photoshop? or something else? Is there a program that allows me to 
> work on every channel (RGB) individually or, even better, can show a 
> "live merge" where I tweak the individual image and the change shows 
> in the merged results on "real time"?
> 4) My main problem right now is, as said, getting the grayscale images

> from every color together into a nice merge, my results are not good 
> enough. I'd welcome any references to online "How to", step-by-step 
> and explanation pages.
>
> Many thanks,
>
> uriel.
> Uriel Trahtemberg
> MD/PhD student
> The Laboratory for Cellular and Molecular Immunology
> The Hebrew University - Hadassah Medical Organization
> Jerusalem - ISRAEL
>
> "God does not care about our mathematical difficulties. He integrates 
> empirically."
> Albert Einstein