This is my implementation of the CBGM. It was designed for testing and changing the various algorithms and is not (therefore) the fastest user-facing package. The idea was that everything be calculated from scratch each time - although in later development I added a Genealogical Coherence Cache for convenience (and speed). If you change the input data then remember to delete the old cache...
- You need these packages installed (these are the Ubuntu names):
- openmpi-bin
- libopenmpi-dev
- graphviz
- python3-dev
- libgraphviz-dev
- Then you can install using pip straight from github, like this:
- pip install git+https://github.com/edmondac/CBGM
- OR clone the repository and do one of the following:
- python setup.py develop # if you want to edit the code
- python setup.py install # if you just want to use it
The main program is cbgm.
Run cbgm --help for info, and for info about the subcommands do, for example,
cbgm combanc --help.
Similarly, see the help for the other programs. They all start 'cbgm_'.
To create the code documentation, run doxygen doxygen.conf
To run the unit tests, run ./test.sh or just python -m unittest discover CBGM
Also see the "examples" folder, including the README.sh in there, which can be executed with bash to test the common CBGM features.
To reference this software, please use the DOI:
