Variable image types for analysis?



I’m suspecting this is a noob question. I have a video file, fura2 imaging with 3042 frames, and an 4 images per experiment (brightfeild, gfp/CGRP, tritc/IB4, and dapi/hoeschtt 33342)

I have figured out how to load the video file from the nd2 format. (following the fantastic tutorials, and a couple of Q and A’s from this team;

But i would also like to load the single image sets into my pipeline. Am i missing something here? Please help

fura2.analysis.cpproj (4.3 MB)

Setting a static ROI over a time series of images

So just to be clear, are all of your channels 3042 frames or just one of them? Either can be done, just the last step will need to change.

You’ll want to configure Metadata so that a) all your channels have a grouping- here I’ve chosen “Well” from the folder name since they all seem to be in the same folder. You also want to turn “Extract from file image headers” on so that you can pull the number of frames in your large movie. (It hasn’t done that properly here since I don’t have your movie, but you get the idea. Also, sometimes the “Extract from image headers” is buggy and you need to save your changes and re-open to get it to work- sorry, we’re working on it!)

If all of your channels are the same number of frames, then the way you’ve configured NamesAndTypes in the pipeline you uploaded is correct- if just the fura is 3042 and the others are single frames, you’ll want to use the Metadata to do the image set matching, so that it knows that they have to all match in the same ‘Well’ but that one of your channels will have another dimension (here I’ve chosen SizeT but you’ll have to look at your extracted metadata to figure out the correct value here). This will allow you to match many timepoints of one movie to a single image in a different channel.


So, only my video file is 3042 images long, consisting of a 340 380 images (see below) (will need to eventually calculate a ratio)

From here i can get to this page, with f2_340, and f2_380 having 3034 frames each. I am struggling with also adding single frame images to it. For example i would like what you have posted in your second picture combined with the picture below.


fura2.analysis.cpproj (4.3 MB)


So to get it correctly configured you still need to
a)In Metadata, add one field that will match across all your images- it looks like they’re all in the same folder (‘Protocol 1’), so you could add another Metadata extraction, where you Extract from folder name
[\\/]Protocol (?P<Folder>.)$
which will give you something that looks like this (now notice there’s a column called “Folder” where all the images have a value of 1)

Then you need to add all your image types to NamesAndTypes- once you do, you can now match by the common ‘Folder’ Metadata, and then for your two channels of fura add the T dimension, like so

Good luck!

Video processing
Cxd Movie analysis - thousands of cells over hundreds of images