Artificial Gene Regulatory Network model and evolution in Python
Recommended installation method uses python's venv module:
git clone https://github.com/d9w/pyGRN.git
cd pyGRN
python -m venv [VENV_PATH]
source [VENV_PATH]/bin/activate
pip install -e .where VENV_PATH is determined by the user (for example, ~/.venvs/pygrn)
pyGRN requires a grns directory and a logs directory. By default, these
will be in the same directory that scripts are run from; for example:
cd pyGRN
mkdir grns
mkdir logsGRN files saved in JSON format will be generated during evolution and placed in
grns, while evolutionary logs will be written in logs.
Examples of usage can be found in regression.py, which can run various
regression problems, and dqn.py, which
uses keras-rl to perform deep Q learning
with a GRN.
To run the tests, use pytest or pytest --cov=./ to get a coverage report.
Certain tests are stochastic and should pass most of the time, but might fail
simply due to random initialization (eg tests/grns/test_grn_compare.py). To
run these tests with verbose output, use:
pytest -v -s tests/grns/test_grn_compare.py
The tests also have many examples of usage beyond the above files.