Installation and Operation

The app is implemented in Python 3 with the help of a variety of packages including flask, matplotlib, networkx, numba, pathos, and attrs. We have released SteadyCellPhenotype as a python package on the Python Package Index. Packages from this repository can be installed using the pip package manager via

python -m pip install steady-cell-phenotype

Commonly, such as on Mac or Linux-based systems, Python 3.x is installed as python3 and you will instead need to run

python3 -m pip install steady-cell-phenotype

This will add two executables to your path, start_scp.sh and scp_converter.py. You can also install SteadyCellPhenotype directly from the github repository, if you would like to contribute to the project, and/or in a virtual environment, see below, which may help with package versioning issues.

Assuming that you are in the top directory of this project. This will open up a web server on the loopback address on port 5000. This means that

• It won’t be accessible to the broader internet and
• On the same machine, you can point your web browser at http://localhost:5000 and access the site.

The flask documentation contains info on how to get SteadyCellPhenotype working for remote users.

Installing from the github repository

Alternatively, one can install SteadyCellPhenotype from the source on github. While in the top directory of the repository, you can install SteadyCellPhenotype's dependencies using, as appropriate, either

python -m pip install -r requirements.txt

python3 -m pip install -r requirements.txt

Then, on Mac or linux, the site can then be started by

./start_scp.sh

You can also install SteadyCellPhenotype as a local python package via pip:

python3 -m pip install -e .

This will install SteadyCellPhenotype as an "editable" package. i.e. Changes made to the source in this folder will apply to the active installation. (Non-editable packages run from a separate copy, created on installation.)

Using a virtual environment

If you encounter compatibility errors between various python and package versions on your computer, it may be useful to create a virtual environment with project-specific versions. (e.g. as of this writing there are issues with the current versions of numba and python 3.9 on macOS.) To create a virtual environment, first find the desired version of python on your machine. (We assume 3.8 below.) Hint: On macOS and Linux, we can use

find / -name python3.8 2>&1 | grep -v "Permission denied"

to find all copies of the python 3.8 executable on the system. Then, using the desired copy of python, run

/whatever/path/to/python/you/found/python3.8 -m venv stdy-cll-phntyp-venv

replacing stdy-cll-phntyp-venv with whatever folder name you desire the virtual environment to reside in. (Often we choose venv inside of the source directory.) All packages which we install will reside in this directory. Then whenever we want to enter the virtual environment, we run

source stdy-cll-phntyp-venv/bin/activate

Now run

python3 -m pip install -e .

The command line tool, scp_converter.py

In addition to the browser interface, SteadyCellPhenotype provides a command line tool: scp_converter.py for advanced users. This tool allows the user to perform various transformations to a MAX/MIN/NOT/polynomial model, including:

• generation of a pure-polynomial model, (i.e. conversion of MAX/MIN/NOT formulae to the corresponding polynomial)
• conversion of some or all formulae to the corresponding continuous version,
• generation of a “self-power” of the system, suitable for computing cycles as fixed points of the composed system.
• generation of several C language programs derived from the model:
  • a simulator which runs the system on a random sample of initial conditions, searching for attractors
  • a simulator which does a complete state space search of the system, searching for attractors
  • a simulator which creates a graph representation of the update function on state space

Each of these generated programs can be compiled using recent versions of gcc and require the header files (*.h) included in the steady_cell_phenotype directory. Output is in JSON format.

Commands line options for scp_converter.py are shown by running scp_converter.py --help