Functions and scripts for processing RNA MERFISH data.
The script run.py is a pipeline that takes the molecules identified by MERlin and a cellpose segmentation and produces a number of quality metrics and outputs, notably a cell by gene table that can be used for further analysis. A scanpy object is also created with an initial clustering using default parameters.
To run the pipeline, a configuration file is required. See config.json for an example of this configuration.
This command will run the pipeline:
python run.py -c config.json -o output_folder -e experiment_name
The output of the pipeline will be put into the specified output folder. The input to the pipeline is defined by the analysis_root, data_root, segmentation_root, and segmentation_name specific in the configuration file, along with the experiment name given in the command. For example, with an analysis_root set to /home/myself/merlin, a data_root set to /home/myself/data, a segmentation_root set to home/myself/segmentation, a segmentation_name set to cellpose_segmentation and an experiment name of exp1, the pipeline will expect the find:
- Raw image files in
/home/myself/data/exp1 - MERlin output in
/home/myself/merlin/exp1 - Segmentation masks in
/home/myself/segmentation/exp1/cellpose_segmentation
Documentation is a work in progress and will be expanded over time. Contact Colin Kern (jckern@ucsd.edu) for any questions.