Migration assay.how to create pipelinesfrom scratch,plz help

Hi there;
I am seriously struggling with creating pipeline for my images. I have been trying over and over since the begging of 2012. Very patient, haaaaa? :unamused: I was very very happy when I saw the video tutorial of construction a pipeline but unfortunately it made more confused. I really need to quantify my data, for a good publication. Here is simply what I am doing;
1- I have DAPI pictures for 3 different cells and the wound is in the centre of the image. The cells are supposed to migrate to the centre and close the wounds.
2- I opened the software to create a pipeline from scratch.
3- Created a folder on my desktop to be my input and out put folder. In this folder images of my 3 cell lines , each has the name of DELP53, partially in it.
4- clicked the + button, Load images , click on exact name matches, write DELP53 in the matched text box.
5- clicked the box (Anlyze images) and never worked … Getting so frustrated now… :blush:
I would really appreciate your help… It is important for me to get this quantification done


When you say that it “never worked”, what do you mean exactly? Did it not do anything or did it produce an error?

Also, it looks to be that if you put in a LoadImages module, followed by a ColorToGray (set to “Combine”) and a IdentifyPrimaryObjects module with the input set to the grayscale image output by ColorToGray, it works reasonably well without any tweaking.
-Mark

Hi again;
Thanks for your help. I figured that out. My problem was my pics were in JPG format. It worked with me smoothly with no problems when I converted them to TIFF.
so, now my question is how can I measure the size of the wound (the hole in the centre of my pics.).I am comparing this between cells.
my assay is simply to create a hole in the centre and then measure how much of the wound is closed by cells as it is a direct representative of cell migration.

So, I loaded my pics as pipeline by adding module (load images and then colortoGRay and then smooth Caucasian filter) after that I added 2 other modules which are identify primary object as tissue and then add the export to spread sheet. every thing works cool till the end when it gives me the excel sheet that I could not understand. to make this easier for you I have attached a copy of a picture and I simply want to measure the size of the hole in the centre.


Hi,

I would suggest creating a mask for the 0hr image, that is, an image the same size as the original which is 1 (or white) within the circle you outlined, and black for everything else. The workflow would then be something like the following:

  • Load the mask using the LoadSingleImage module, along with the timepoints loaded using LoadImages.

  • Use MaskImage with the original image and the mask as inputs to remove everything outside the circle. Use IdentifyPrimary (along with Smooth beforehand if needed) to detect the cells within the circle.

  • Use MeasureImageAreaOccupied to measure the area occupied by the cells.

  • Use MeasureImageAreaOccupied to measure the area occupied by the circle.

  • Use CalculateMath to divide the cell area by the circle area.

This should work contingent on whether the wound is stationary in (X,Y) location from one timepoint to the next. If not, you should work on positioning your sample in order to make it so.

Regards,
-Mark

Ok here is what I did ;
I did exactly what you told me and it keeps giving me an error. So, I thought to create a pipeline which has both pictures (0hrs and 16hrs).
+module: Load images.
+module: single image
+module: clortoGray
+module: smooth “Cacausian filter”
+module: identify primary object from filtered images “cells”
+module: mask images “origBlue”
+module: measuare the area occupied " total area by cells" was very confused here but that’s what I chose
calculate math: divide and again confused to how I should do it. many options that I was not be able to pick one easily.
End results: no data, Error while processing. I tried so many different options within each module and trust me that I am sending this when I almost lost hope to do it all by myself. I know that it sounds very simple for you but it is very complicated for me :unamused: .
So, I really wish if I can send you these 2 pictures only and you do it and then I can use the exact same criteria for my other samples. please I seriously fed up and know that it should work with no such headache.
thank you

here is the second picture; I deeply appreciate your help

Hi,

Sorry for all your troubles. First, as I say in the FAQ,

Would you please help us help you, and post your existing pipeline file (*.cp)? Even if it errors.

Thanks,
David

Oh, and have you seen and run successfully our “Wound Healing” Example Pipeline here?
cellprofiler.org/examples.shtml

This should run at least!
Cheers,
David

hello there;
I am very very sorry for the inconvenience. I was on a trip for the holiday and then I lost my laptop with all the data on. Thanks God I have saved the data on the school’s network. to answer your question, yes i did look at the assay you and I was very excited because it says measuring the covered area in wound healing assay. but it did not have any .cp or .mat files it is only 2 JPEG and 1 text file.

The wound healing ZIP file should have 3 files: 2 JPGs and a .cp file (which is text). I’m posting the ZIP file for your use, along with the .cp file just in case.
-Mark
ExampleWoundHealing.cp (5.87 KB)
ExampleWoundHealingImages.zip (1.11 MB)

thank you. I will check it out and mimic the “optimizing points”.