Identifying cell migration in a bulls eye experiment


#1

Hi,

I am relatively new to CP and am having quite a few difficulties.

I have done a cell migration assay and now I would like to calculate how much my cells migrated out of the initial drop. So I would like to calculate the surface area occupied by the cells and the feret diameter.

I thought I would identify the drop+migration area as a single primary object and then measure that, but I cannot get it to identify the object. Is there a better way to do it?

Here is an example of an image I am trying to analzye:

https://drive.google.com/open?id=1tgciZjXxxiByapSp3wAGxki3i2UXj-pT

Many thanks,

Giulia


#2

Hi,

It might be useful if you upload the pipeline you have so far so we can see how far you’ve gotten and where exactly the problems are.

Some functions you might find useful to test out are the expand ones of “ExpandOrShrinkObjects” module and the “MergeObjects” module.

Best,
Laura


#3

Giulia, I bet you will want to identify the cells using IdentifyPrimaryObjects and also identify the drop using IdentifyPrimaryObjects and also identify the migration area (probably using IdentifySecondary, building out from the original drop shape - then IdentifyTertiary to subtract the drop part away from that).

Then you can use something like RelateObjects (or maybe Mask, not sure) to associate cells that overlap the migration area. MeasureAreaOccupied could then help you.


#4

Hi,

thanks of the useful input. I had something in mind like that but I am failing at identifying the primary object. I tried playing around with the parameters and putting as input both the raw image and the mask/threshold.
Here is my pipeline. migration.cpproj (399.7 KB)

I even thought of just measuring the area covered by the cells rather than just the migration as I have images of time 0 (do just the initial drop) so I could upload those images and then subtract the area covered by the drop to the total migration area.

I am new to CP so I might be missing something really obvious, sorry!


#5

Hi,

I think a big problem is just that your images are HUGE. You might want to try to do a Resize module to way downsample the image in order to make it small enough for your computer’s memory to handle.

As far as accurately finding the cells, you may want to try something like ilastik, which you can use on its own or in combination with CP.