SAMA documentation - How to use SAMA?

Overview

In imageJ, go to Plugin > Sama. There are 2 main components available – SAMA-images and SAMA-analyze.

Sama-images performs the image analysis part, and produce quantitative data for every image stack. The processing is modular, for example, the user may want to do only Tier 1 with a few options, and compute other options or tiers later.

Sama-analyze gathers the output of sama-images and enables to represent, analyze and export it. The analysis includes both treatment effects and the comparison between the different replicates to check the robustness of the results.

Example dataset

The simplest way to first use SAMA and check your installation is to download the sample set of images available here  and follow the steps below.

• Unzip Example_SAMA.zip.
• In Plugin > Sama, run SAMA-images and uncheck the lumen options in tiers 1, 2 and 3. Only Basic morphometrics and Branching measurement should be checked.
• In the "choose source directory" dialog select the denoised folder in the example_SAMA folder.
• Wait for the processing to finish.
• In Plugin > Sama, run SAMA-Analyze. Uncheck Lumen and hit ok.
• Enjoy the pdfs produced in the folder SAMA-analyze output. The main results are in the coarseplot pdfs (the different pdfs correspond to different filters on the structures, see below).

Folders

Important: create a folder where the analysis of the dataset will take place, and put all the source image files in a subfolder of this main folder.

SAMA will create other subfolders as follows:

• Sama-object map (with image stacks where all structure have a color corresponding to its label).
• Sama-prelumen images (the images preprocessed for lumen analysis)
• Sama-tree analysis (the skeletonized images)
• Sama-Image data (the various raw output sheets)
• Sama-analyze output (the pdf and reported sheets).

The folders Sama-object map and Sama-tree analysis can be used to check the validity of the image analysis.

Type of images to be analyzed

SAMA supports any file format supported directly by fiji.

There is no intrinsic limit to the size of the original image but the memory available to Fiji. Note that you can change the later. Also in case of tiling, the stitched image should be used.

SAMA takes into account the scales that can be set in Image >proprerties. In particular:

• The filters by size are in unit³ (for example µm³, pixel³=voxel, …). The default of 500 is reasonable in voxels.

• Relevant results will follow these units.

• Thickness analysis requires that the scale is the same in the three direction to provide relevant results.

SAMA is designed to analyze a population of structures and not a few structures at high magnification. In the later case, the patterns generated by cells may disturb the result.