search

fish-quant / big-fish-examples

Notebooks to illustrate how BigFISH is working.
BSD 3-Clause "New" or "Revised" License
13 stars 8 forks source link

"voxel_size" and 'spot_radius' parameters #14

Open ZhijianZhou01 opened 10 months ago

ZhijianZhou01 commented 10 months ago

Dear fish-quant development team,

Recently I've been learning how to use bigfish which is a really great software. However, I don't have a thorough understanding of the meaning of two parameters in bigfish although help documentation is provided, and I look forward to your help.

(i) In the "detection.detect_spots" function, what do "voxel_size" and 'spot_radius' parameters mean, and what is the effect of the size of their value on the number of points detected? Besides, the "voxel_size" parameter also appears in "multistack.detect_spots_colocalization" function.

(ii) How do I change the background color of the output image, such as black? I tried to make changes to the source code of the drawing, but it didn't work, such as plt.rcParams ['axes.facecolor'] = 'black' or plt.figure(facecolor = 'black').

I look forward to receiving your help in time to make my experiments work.

Yours sincerely, Zhi-Jian Zhou

Henley13 commented 10 months ago

Hi Zhou,

i) Parameter voxel_size is the actual size (in nanometers) of the surface/volume represented in your voxel. It's related to your experimental setup. Parameter spot_radius is the expected radius of your spot (in nanometers). Based on this information, the algorithm can calibrate the detection and choose an appropriate distance to discriminate two adjacent spots.

ii) You should set the colormap of the plot, not the facecolor. Maybe you can try to change the cmap parameter in plt.imshow() with something like cmap="greys". See for example: https://stackoverflow.com/questions/9638826/plot-a-black-and-white-binary-map-in-matplotlib

Best regards, Arthur

ZhijianZhou01 commented 10 months ago

Thank you for your generosity and patience, I have another question, does the threshold in this function refer to the threshold of diameter (nm) used to identify the expected point?

spots, _ = detection.spots_thresholding(image_log, mask, threshold)