hernandomv / cellprofiler_analysispipelines

For RNAscope

This is a pipeline to visualize and analyze the results of CellProfiler on RNAscope experiments.

1. Take images in the SP8. Individual or tiled, doesn't matter. Do a z-stack and in Fiji make a maximum projection.
2. Split the channels and save them individually, with their default Fiji-titles (But removing spaces in the name!), in the folder where to do the quantification.
3. Copy the Cell-profiler project file in the CellProfilerProtocols folder into the same directory.
4. Modify the project to work with your images, and change the two output saving paths.
5. Run it. It should produce a .csv file with the statistics, and an overlay image summarizing the results.
6. Run jupyter notebook.

For Rabies-cFos quantification

This pipeline aims to quantify the relative (to the full image) c-Fos staining in each cell infected with rabies.

step 1

Take confocal images (current specifications below) of your rabies channel and your c-Fos channel. Resolution: 0.9240 pixels per micron Pixel size: 1.0823x1.0823 micron^2 Objective used in SP8: 10x AIR Bits per pixel: 8 (grayscale LUT)

step 2

Run CropLIF.ijm in FijiMacros. This will: -Crop your images (pair always the two channels) to the region of interest and save the channels separately using this format: MouseID_StarterCells_cFosCondition_SlideSliceNumber_SideoftheBrain_channel.tif (channel being 'cFos' or 'rabies'). -Save them in a specific folder

step 3

Run IlastikProjects/RabiesContentQuantification.ilp and load your ...rabies images. Check that it classifies it correctly and retrain if needed. Save the outputs in a separate folder (e.g. IlastikOutput/) following these guidelines: https://github.com/CellProfiler/CellProfiler/wiki/How-to-use-Pixel-Classification-in-CellProfiler

step 4

Run FijiMacros/GroupPercentileThresholding.py in Fiji to threshold the cFos channel.

step 5

Run CellProfilerProtocols/RabiesContentQuantification.cpproj to get the tables. Here, drag the main folder to the Images field (first), and specify which percentile is low, med and high (NamesAndTypes) Change the input and output folders to the main directory in 'View output settings'

step 6

Activate conda environment imageanalysis [TODO: create requirements] Run jupyter notebook