gemmr - Generative Modeling of Multivariate Relationships

gemmr calculates required sample sizes for Canonical Correlation Analysis (CCA) and Partial Least Squares (PLS). In addition, it can generate synthetic datasets for use with CCA and PLS, and provides functionality to run and examine CCA and PLS analyses. It also provides a Python wrapper for PMA, a sparse CCA implementation.

Hardware requirements

GEMMR runs on standard hardware. To thoroughly sweep through parameters of the generative model a high-performance-computing (HPC) environment is recommended.

Dependencies

• numpy
• scipy
• pandas
• xarray
• netcdf4
• scikit-learn
• statsmodels
• joblib
• tqdm

Some functions have additional dependencies that need to be installed separately if they are used:

• holoviews
• rpy2

The repository also contains an environment.yml file specifying a conda-environment with specific versions of all dependencies. We have tested the code with this environment. To instantiate the environment run

>>> conda env create -f environment.yml

Installation

The easiest way to install gemmr is with pip:

pip install gemmr

Alternatively, to install and use the most current code:

git clone https://github.com/murraylab/gemmr.git
cd gemmr
python setup.py install

Installation of gemmr itself (without potentially required dependencies) should take only a few seconds.

Documentation

Extensive documentation can be found here.

The documentation contains

• Demonstration of the gemmr's functionality, including exptected outputs (all of which should execute quickly)
• Juyter notebooks detailing generation of the figures for the accompanying manuscripts
• API reference

To generate the documentation from source, install gemmr as described above and make sure you also have the following dependencies installed:

• ipython
• matplotlib
• sphinx
• nbsphinx
• sphinx_rtd_theme and run (in the doc subfolder):

make html

and open doc/_build/html/index.html .

Citation

If you're using gemmr in a publication, please cite Helmer et al. (2020)