ODE_MCMC_tool provides a C++ adaptive Markov chain Monte Carlo (MCMC) and parallel tempering MCMC framework for ordinary differential equation models, allowing Bayesian estimation of model parameters and/or initial concentrations.
The software has currently been tested with Linux (RHEL 7) and Mac OS X with MacPorts (without the static compile flag). The software requires the Ceres Solver, Eigen3 and Boost C++ libraries (including the development header files). The build system currently uses SCons.
Compiling the C++ code
To compile the software go into the main directory and run the command:
Where the -j option specifies the number of simultaneous tasks you wish to use to speed up compilation.
Running MCMC on the example data
Some example ODE models and real experimental data are provided from some Bacillus megaterium cell-free transcription and translation experiments.
Go into the example_data/bmeg/xylose_xylr or example_data/bmeg/competition_experiment directories and follow the example command lines provided in the README.md files.
This software is distributed under the GNU GPL license, version 3.
(C) James T. MacDonald, 2017. Imperial College London.