The goal of starfish is to prototype a reference pipeline for the analysis of image-based transcriptomics data that works across technologies. This is a work in progress and will be developed in the open.
See this document for details. The diagram below describes the core pipeline components and associated file manifests that this package plans to standardize and implement.
Starfish supports python 3.6 and above. To Install the starfish package, first verify that your python version is compatible. You can check this with pip, which may be called pip or pip3 depending on how you installed python.
The output should look similar to this:
% pip3 --version
pip 10.0.1 from /usr/local/lib/python3.6/site-packages/pip (python 3.6)
While not required, you may wish to set up a virtualenv. To do this, execute:
% python -m venv .venv
Install the starfish module in edit-mode and all the dependencies for starfish:
% git clone https://github.com/spacetx/starfish.git
% pip install -e starfish
See the notebook for a fully worked example.
You can also re-produce the notebook results with the command line tool. For usage, run starfish --help
Running the commands below will re-produce notebook results.
mkdir -p /tmp/starfish/raw
mkdir -p /tmp/starfish/formatted
mkdir -p /tmp/starfish/registered
mkdir -p /tmp/starfish/filtered
mkdir -p /tmp/starfish/results
python examples/get_iss_data.py /tmp/starfish/raw /tmp/starfish/formatted --d 1
starfish register -i /tmp/starfish/formatted/experiment.json -o /tmp/starfish/registered fourier_shift --u 1000
starfish filter -i /tmp/starfish/registered/experiment.json -o /tmp/starfish/filtered/ white_tophat --disk-size 15
starfish show /tmp/starfish/filtered/experiment.json
starfish detect_spots /tmp/starfish/filtered/experiment.json /tmp/starfish/results dots --min_sigma 4 --max_sigma 6 --num_sigma 20 --t 0.01
starfish segment /tmp/starfish/filtered/experiment.json /tmp/starfish/results stain --dt .16 --st .22 --md 57
starfish gene_assignment --coordinates-geojson /tmp/starfish/results/regions.geojson --spots-json /tmp/starfish/results/spots.json --output /tmp/starfish/results/regions.json point_in_poly
starfish decode -i /tmp/starfish/results/encoder_table.json --codebook /tmp/starfish/results/encoder_table.json -o /tmp/starfish/results/decoded_table.json iss
To see an interactive web-visualization of the final decoded result, run the following commands
- Install nvm if you don't have it
- Install node
nvm install node - Install budo
npm install budo - From starfish/viz run
npm install - From starfish/viz run
npm start
To use the interactive notebook widgets, run the following commands. Please note that the widget currently supports Jupyter notebooks, but not Jupyter Lab.
pip3 install -r REQUIREMENTS-NOTEBOOK.txt
jupyter nbextension enable --py widgetsnbextension
Clone the repo and look through (CONTRIBUTING.md)[CONTRIBUTING.md]
% git clone https://github.com/spacetx/starfish.git
% cd starfish