Run the code with python main.py <config filepath>. For example
python main.py config/default.yml. The output will be saved under
an experiments folder, in a subdirectory with the same name as the
config file. The config file config/default.yml has a hyperparameter
configuration that we have found usually gives reasonable results.
We use a particular format for the CSV referenced in the configuration files.
Required columns can be found in dfconst.py. If you use a different set
of column names for your dataframes, it should be straightforward to use
them by modifying dfconst.py.
Spherical Phenotype Clustering. Luke Nightingale, Joseph Tuersley, Andrea Cairoli, Jacob Howes, Andrew Powell, Darren Green, Amy Strange, Scott Warchal, Michael Howell. bioRxiv 2024.04.19.590313; doi: https://doi.org/10.1101/2024.04.19.590313