(part 1) Processing parameters
The parameters dialog window for the batch processing looks like:
These are the main parameters' values that were used in our publication.
If you want to adjust them for your dataset, please read their description below.
Additionally, there is a set of implicit "hardcoded" parameters in each macro,
You can find them on the corresponding (steps) description pages.
represents standard deviation of the point spread function (PSF) of your microscope
approximated with a Gaussian. It does not have to be super precise.
A good estimation could be obtained by drawing a line ROI intersecting BB ring in a single plane,
and build intensity profile (Fiji->Analyze->Plot Profile or Ctrl + K).
In the new plot window choose Data->Add Fit->Fit Function:Gaussian, see below.
Final fitting parameters will be displayed in the Log window, you need to take the value of the parameter d.
Notice, that it is provided in the scaled units (micrometers in this case).
This parameter is used to calculate minimum diameter, when it is not possible to observe a hole in the simulated cross-section of a BB.
This parameter estimates maximum possible diameter of BB cross-section ring.
Again, it is provided in scaled units.
It can be quickly estimated by inspecting the data and taking a bit higher value.
During BB detection the macros will make a series of circles starting with max diameter and shrinking with this step.
The circles images will be used to find position and diameter of BB ring cross-section at different planes.
Smaller values of this parameter provide more precise detection at the expense of longer computational time.
Again, it is provided in scaled units.
NB: this parameter is specified in pixels (XY pixels).
Since stained proteins could localize outside the BB ring, we need to decide, how much area around BB we should extract.
This parameter specifies the diameter of the tube around the BB centerline, that is going to be extracted
from the original volumetric data.
So it is defined by the "furthest from the center" possible localization of proteins of interest.
A rough initial estimate could be given by 2*maximum diameter / XY pixel size.
These factors allow scaling down pixels size of the volume in the specified axes (upscaling the data).
This will happen after BB detection, straightening, but before localization
of the basal feet and rotation registration (and averaging).
Upscaling should improve averaging, allowing more precise registration.
This parameter represents the maximum number of iterations during consecutive averaging/registration.
From our experience, on our data the value of 4 works pretty good, higher iterations do not improve the outcome.