Hello,
I am new to microscopy and have recently executed my first experiment using RNA FISH probes to target various vRNA segments of IAV within A549 cells. The virus has been engineered to express GFP and the probes are conjugated with an Alexa Fluor 647 dye. I used confocal microscopy to image my samples. I am using flow in parallel, but it would be optimal to quantize the number of viral particles via microscopy. I am having difficulty in finding a workflow by which I can essentially ask ImageJ how many cells express GFP, and of the ones that do, what is the number of Alexa Fluor dots (~ viral particles) within the cell?
How might I go about this? Any and all help would be appreciated. I attached an image to illustrate the question.
Image J measure intensity. Based on your figures, select the area using tools and press measure key. Do it for control as well. Copy down the reading and understand to make your graph by calculating the statistics and p-value. good luck.
you will have to choose several similar representative images for same data set. Then select 1 cell at a time by draw tool...and measure...
i think...but i am not sure that in imageJ, free version...there is option to select wavelength....you may select the exact colour/fluorescence, witch you want to quantify, in some other image quantifying softwares...hope you must have received one such software with the microscope that you are using....else just ask the customer support of your microscope, stating the model and wavelengths you are using/want to quantify..
I hope this helps!! do let me know!!
Open your pictures through "stack"
Split to the different wave lengths by "stacks- stack to image" or image-colour-split channels
Choose the GFP channel picture, stand on it and determine the threshold for GFP by "image-adjust-threshold"
Analyze – include holes- check V, in order to include areas in a GFP cell that do not display the GFP
Check v also on the options : display results, clear results, summarize, add to manager,
And most important check v on the "show: masks"
You will get the mask in a separate picture, change the colour of this picture from black on white background – to white signal on dark background by "image-lookup table- grays"
Then in order to analyze the 647 signal in the masked GFP cells:
process: image calculator: image 1 –choose your Alexa Fluor 647,
operation: choose, and
image 2: choose mask
check v on the "create new window"
you will get a new picture with your 647 signal – only in cells with GFP
now that you have the 647 picture versus the 647 in GFP cells, you can get the total signal versus the signal inside the GFP area by using the "analyze particles" or "measurement".
The "analyze particles" analysis is great for your viral particles, look into the different options - size, number, intensity..
Thank you Tziona Ben-Gedalya
I have implemented your workflow, but had included some other steps to try to increase efficacy and reduce bias, which I would greatly appreciate your opinion on.
Prior to processing the 647 signal in your provided workflow, I have decided to attempt to reduce noise, deconvolute my puncta, and decrease bias. I would greatly appreciate your thoughts on what I have done.
I have additionally thought about generating a point spread function image to deconvolute my puncta, but haven't implemented this yet. I am curious if this workflow sounds appropriate for what I am trying to do. I have noticed my particle count can change dramatically depending on the threshold algorithm I choose. If you have any further recommendations it would be greatly appreciated!
for general background and noise, you can also try:
process- noise-despeckle
Process - Subtract background – rolling ball 50 pixels, uncheck the light background
of course anything you use- use exactly the same conditions on all samples treatments and controls.
The right way to determine a threshold is to have controls that were prepared together with the experiment and using exactly the same process, just without the specific dye or probe, or treatment- without the virus
your images seem to be in saturation, take unsaturated images in order to be able to quantitate the signal.
good luck with your work!
Tziona Ben-Gedalya Thank you very much for your answers they have been very helpful!
Hi dear Draven Rane
I understand you want to determine pink particles number or area.
I can suggest you "color separation technique"
1) go to
Image -->Color--> Channel tool
2) just mark channel 1 (red type)
It will increase the resolution of particles
3) duplicate from that picture ( Only from channel1)
4) go to
Image --> Adjust --> threshold
5)for separation between particles
Process --> Binary--> Watershed
(I prefer manually separation with white color)
6) Go to
Analyze--> Analyze particles
:)
Good luck
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis