Hi, everyone!

I've been trying to use FiloQuant in Fiji (as described here: https://rupress.org/jcb/article/216/10/3387/38936/FiloQuant-reveals-increased-filopodia-density) to analyse the length and density of filopodia in RasV12-expressing clones in the wing imaginal discs of Drosophila melanogaster larvae (acquired by confocal microscopy). Using the single-image option of FiloQuant, I've been able to set a proper threshold for the cell body, but not for the filopodia themselves. And when I get something in the final result, the detected filopodia are often broken (e.g., instead of recognizing the entire filopodia that I see with my own eyes in the original image, it only recognizes bits of it). I have tried some things:

-> I've tried to used the GFP channel only (Ras-V12 clones are marked with GFP);

-> I've tried to use the phalloidin channel only (but it gets a bit tricky to minimize the signal from neighboring cells, so I have filopodia being "detected" everywhere");

-> I've tried to use a merge of these two mentioned channels, and I get the same issue;

Do you think I should use a specific file type (e.g., .tiff instead of .czi) or some type of image preprocessing first (e.g. to increase the intensity of the signal/ contrast, so the filopodia are easily recognizable; or to binarize the image first and then apply the plugin)? Also, does FiloQuant work well with 20x images, or is it better to use 40x images instead?

Thank you in advance,

Marta