More than 3 channel analysis

Hi There,

I have four separate channel of individual biological importance. One of them is DAPI. I want to segment Nuclei as Primary object and quantify intensity in remaining three channels. I don’t know how to call/label four images for further processing. Are we only limited to process 3 channels using RGB conversion?

Following are my steps. I need help in step 2

Step 1: Converted to Well, Position, and plate metadata using following RE.

–W(?P[0-96]{1,2})–P(?P[1-8]{1})–Z0–T0–(?P.*).tif

(?P[0-999]{3})

Step 2: Now I want to call Channel 1 as my nuclei and perform primary object (I don’t know how to choose “ChannelName” within “IdentifyPrimaryObjects” ruleset)?

Best Regards,
Deven

Hi Deven,

  • If your channels are in separate files, you’ll want to use NamesAndTypes to associate all your channels together (the help for that module is a good place to start).

  • If your channels are all in one file:

    • If they’re actually RGB tifs (an easy way to tell- if you open them in ImageJ/FIJI does the header it say “(8/16) bit” or " ((8/16) bit) RGB"; also if it’s RGB you WON’T get a channel slider on the bottom of the image in ImageJ/FIJI) then you can’t actually cleanly separate your 4 separate channels, you’ll need to go back to your microscope and export them as a non RGB tif

    • If they’re individual channels in one file (again, the “channel slider” if you open them in ImageJ/FIJI is a good hint that this is what you have) you can load them as “Color Image” (rather than “Greyscale Image”) in NamesAndTypes and use the ColorToGray module* in Split->Channels mode to split out your different channels. You can then pick whatever one is your DAPI for IDPrimary.

(*=Note that there’s a display bug in CP 3.0 right now where this won’t work for 4 channel images if you have the “eye” open next to the module; make sure to close it if you’re using 3.0, and we’re hoping to have that fixed soon!)

1 Like

Hi @bcimini, I am trying to perform a similar analysis as Novel_Spirit on a 3D series of images. I have another thread open at the moment to get my tracking of Primary Objects (nuclei) working. But the quantification of fluorescence intensity in the remaining 3 channels will be the next step.

May I ask you for an example of a pipeline that uses the Primary Objects identified in one channel to apply them to other images, with corresponding Z positions, and then to quantify from them?

Thanks a lot and best wishes,
Sophie

Hi @Novel_Spirit,

Hope you are able to get the channels properly. Another tip to add to @bcimini suggestions.
If your channel information is available in Metadata you can get it in “NamesAndTypes” by set the match rule “metadata” “Does” “Have Channel matching”. You could check this information from FIJI/ImageJ Image>show info or using Bio-formats importer.
Hope this helps.

Regards,
Lakshmi
Fujifilm Wako Automation (Consultant)
www.wakoautomation.com
For CellProfiler training or optimised pipeline write to,
lakshmi.balasubramanian.contractor@fujifilm.com

Read more on our site.
Yokogawa CV8000 - The Ultimate in Confocal HCS
https://www.wakoautomation.com/products/yokogawa-high-content-imaging

Join us at, https://www.slas2020.org

Hi @Sophie_26,

Though there is no direct example pipeline This could be achieved in two ways.

  1. Single pipeline: you will have to input all the channel images. Once you identify nuclei, you can compute the intensity measurement for the pixels of nuclei region from other channel. This could be computed in the “MeasureObjectintensity” where you choose your channel of interest in image to measure & “objects to measure” as your nuclei. At a given point of time you could use more than one image as well. In your Z-stacks, it would measure from the corresponding channel for every frame.
  2. Two pipelines (mostly): One pipeline you have the nuclei segmentation & save is as mask file i.e as mutiframe .tif with the binary masks of the nuclei. In the second pipeline, along with your other channel as input image, load this binary masks and proceed to compute measurements as mentioned above

Hope this link too helps!!!

Regards,
Lakshmi
Fujifilm Wako Automation (Consultant)
www.wakoautomation.com
For CellProfiler training or optimised pipeline write to,
lakshmi.balasubramanian.contractor@fujifilm.com

Read more on our site.
Yokogawa CV8000 - The Ultimate in Confocal HCS
https://www.wakoautomation.com/products/yokogawa-high-content-imaging

Join us at, https://www.slas2020.org

Hi Lakshmi,

Thank you very much for all these suggestions! I’m currently trying to implement them with some help from our imaging department. Something certainly will work!

Best,

Sophie