C. elegans pharyngeal bulb fluorescent intensity

Hi all,

Thank you so much for the Worm Toolbox, it’s wonderful!

I’m also attempting to measure fluorescent (GFP) intensity in the pharyngeal bulbs of my worms. The point of my project is to see if polyphenol antioxidants will prevent a GFP flare up- so my question: how can I identify bulbs that aren’t lighting up with GFP? The bulbs are very difficult to see in the picture by eye. Although this is great and means my experiment is working, it makes quantification of a nonexistant fluorescence very difficult.

Any ideas would be greatly appreciated! If not, I’ll stick with qualitative data analysis…

I’m quite confident it should be possible to quantify your phenotype using the Worm Toolbox.
There are a few alternative approaches:
A. If you can use a pharynx-specific marker, use IdentifyPrimaryObjects on this channel and measure intensity in GFP for pharynx object.
B. If you do not have a pharynx marker, you can still extract fluorescence from the approximate position of the pharyngeal bulb by using the Worm Toolbox to untangle clustered worms, ‘straighten’ the worms, and map them to a common low-resolution atlas.
First of all, you need a good foreground/background segmentation of the worms, if possible using a bright-field image of the worms.
Using non-touching worms (selected by size-filtering or EditObjectsManually), use the UntangleWorms module, in the ‘Training’ mode, and train a model of the variability in shape of your worms (the .xml file produced can be used for all your experiments, as long as you use the same microscope settings & worms). Thereafter, use the UntangleWorms module in the ‘Untangle’ mode, followed by ‘StraightenWorms’, where you specify the GFP channel as the image to straighten, and specify ~6 transverse segments (atlas sub-regions) from which GFP intensity is extracted. Without a common head-marker, you may not know in what end of the worm the pharyngeal bulb is located, but e.g. using the difference in GFP fluorescence between sub-segment 2of6 and 4of6 you should be able to separate the two phenotypes.
We will soon publish a number of sample pipelines on the website, and you’re welcome to send a sample image so that I can help you set up your pipeline.


Wow, thanks for the help! The only challenge with your second suggestion is that for this part of my research I’m only analyzing one worm at a time, because its difficult to see the GFP at anything less than 40X. I’ve posted a picture of my positive control (the super green worm) and my normal control. Like I said, the hard part is quantifying a pharyngeal bulb that’s hardly visible, like in the normal control picture.

If you include a pipeline for this in your tutorials that would be wonderful! If not, thank you for taking the time to respond!

The WormToolbox has been developed mainly for low-resolution data where the whole worm fits within the field of view. It is not a problem that there is only one worm per image (that only makes things easier), but for the Untangling-Straightening-FeatureExtraction to work, you have to use slightly lower magnification and also acquire a bright field image that can be used to define what is worm as compared to background, which is difficult to do from your fluorescence images alone.