A Python package to perform Bayesian inference on enzyme kinetics data.
Currently, the package implements custom Theano-operators to perform inference both on steady-state data of enzymatic systems and time-dynamic data. These operators can be used together with PyMC3 and implement full gradient support for the usage of Hamiltonian Monte Carlo methods.
The current supported version requires a working installation of Amici. This requires the following Ubuntu packages:
sudo apt install libatlas-base-dev swig
sudo apt install libhdf5-serial-dev
As well as the following Anaconda installation and environment variable:
conda install -c conda-forge openblas
export BLAS_LIBS=-lopenblas
The BayERN package can then be installed via pip:
pip install git+https://github.com/mgbaltussen/[email protected]
Developed by Mathieu G. Baltussen in the Huck Group, Department of Physical-Organic Chemistry, Institute for Molecules and Materials, Radboud University.