I know if I want to calculate transfection efficiency it would be done by this equation LIKE THIS
% Transfection efficiency = (number of cells stained with fluorescent positive control dye/total number of cells per field) X 100
But I had the ability to count positive green transfected cells with GFP plasmid manually through image J but I cannot define the borders and count non positive cells so is there away to recognize these cells (Magnification 20x)!!
If you really want quantitative data on which cell is positive and which cell is negative you can use flow cytometry but I think it is too complicated for your purpose. So just simply counting manually is the best way to go probably.
Dear Doaa Yamani ,
As mentioned above FACS might be the way to go to obtain best results. However, there are a few things you can try with the imaging data you already have.
Option 1 - segmentation approach:
Task 1: segment the fluorescent cells. As in general there is not much overlap between them this should be fairly easy by just taking a fluorescence image, applying gaussian filter to improve quality and remove some noise, and then some segmentation, probably Otsu’s method is enough.
Task 2: segment all cells. You can try this on transmission images, phase contrast or DIC. However, this might be complicated due the contrast quality of the images, and in general you have large confluency.
A better way to solve task 2 might be to use a nuclear staining to make cell segmentation easier.
By completing task 1 and 2, then you can calculate your transfection efficiency.
Option 2 - quick and dirty calculation
Another trick you can try if you are only interested in ball-park figures is to first calculate the confluency of your sample. Let’s say 70%. Then calculate the confluency of your fluorescent images. Let’s say 30%. The you can estimate your transfection efficiency by:
Cells per image = Confluency(70%) * image_size / cell size;
Fluorescent cells per image = Fluo_Concluenfy(30%) * image_size / cell size;
Transfection efficiency = (Fluorescent cells per image / Cells per image) * 100;
Transfection efficiency = (Fluo_Concluenfy(30%) / Confluency(70%))*100 = 43%;
Note that you do not really need to know the cell size if you assume that they are all similar and that you take a lot of images to make the statistical significance good.
Best,
- Badges
- Science method