I’m trying to do a fret ratiometric analysis in time-lapse images for cells that change their shape along the time. Most of the signal of interest is located at the membrane of the cell, so I need to keep track of the membrane and segment. However, is difficult to do with typical segmentation because of the mean intensity changes along time. Do you have any ideas to make it? I have images with 4 channels, and I think the 4th is the easiest to segment.[Uploading: Teste.tif…]. I have attached a test file.Teste.tif (3.2 MB)
It looks like your pressed the post button before your file was finished uploading. Maybe you could try again?
Please take a look at the ‘Hwada’ post in this forum.
They have a plugin which does exactly what you are asking for straight out of the box called AutoCutB, with a couple of other goodies you may very well find useful as well.
When you install the plugin it will list as “Hwada” so it may be difficult to find in the list at first.
I have a question.
What did each channel observed?
(e.g. ch1:DIC, ch2: CFP, ch3: YFP, ch4 CFP->YFP)
And our lab. reported a following paper, please refer it for FRET analysis.
A Switch-like Activation Relay of EGFR-ERK Signaling Regulates a Wave of Cellular Contractility for Epithelial Invagination.
In this paper, we displayed the FRET ratio image with cell shape by IMD.
If your segmentation meaning is separate the signal of FRET ratio, please try to display by IMD.
I think IMD is robust against intensity change compared to just ratio calculation.
I upload the processed your sample image.(2ch and 4ch)
FRETratio_CalcNo3_min0.25_max1.2.tif (403.7 KB)
In addition, I update the plugin “AutoCutB”.
The old version had 2 problems.
First is methods are swapped.
Second, if the intensity is shifted to a higher position like your image, the Mod method can not detect the target position.
I fixed these points.
Please get new version “AutoCutB”, and try it for your image.
I uploaded a following plugin.
It is used in the paper that I introduced earlier.
If you are interested please try using it.
I am sorry,
The previous file has some debug process.
So if someone downloaded it, please download new file again.
thank you for your reply. In the image that I uploaded the channels of interest are ch2 (donnor) and ch3 (fret), the others may be used for segmentation. I tried to get the contour of the cell with ch4 and the older AutoCubB seems to work fine, but I’ll try the new version.
thanks for the plugin. I was reading the paper and try the plugins but I cannot have the same result. I cannot have the values of my ratio between 0.25 and 1 with LIMD. the output of the image change to 8 bit so I am assuming that is why I cannot have the same pixel values.
Also I tried to check every combination with autocutB, I could not find a step by step instruction in the website so I have been trying all the possible combination.
Please see attached the image that I am trying to analyse. Is there any problem with my process?
cfp yfp nucleu.tif (4.0 MB)
Thank you for trying.And I tried to test your image.
I recommend setting Min and Max values to 0.6-0.8 and 1.3-1.5.
And I will write a simple process chart.
- open the image
- run this plugin
- select the method “Like IMD”
- set Min Ratio and Max Ratio
- push Calc button
- change the LUT to LIMD2
- retry from 4, if the created image is not good
thank you Hwada, I will try again.
Thanks again for the plugins