revreaddy

This project is not maintained. I strongly recommend https://github.com/readdy/readdy

A particle based reaction-diffusion simulation with a reversible integrator, which obeys detailed balance. Revreaddy is mainly a C++ software which is wrapped to be accessible from python. Main purpose of this software is to try out different reaction-diffusion schemes.

This version features:

• Brownian Dynamics integrator with optional Metropolis-Hastings correction
• Reactions that allow particle creation/destruction e.g. Fusion reaction A + B <--> C, with forward and backward rates
• customizable particle types with different radii and diffusion constants
• two possible repulsion potentials, harmonic repulsion and Lennard-Jones interaction
• geometric building blocks implemented as single particle potentials, currently repulsive plane "wall" and a double-well potential in one dimension
• cubic periodic boundary conditions, can be switched off (not recommended)
• observables which are calculated on the fly and saved to binary (.hdf5/.h5) files or text files (.dat/.txt). These are acceptance, energy, mean-squared-displacement, probability density, radial distribution function, trajectory
• Neighborlattice force calculation leading to overall complexity O(#particles) (exception when rdf is calculated on the fly)

Dependencies

Linked libraries:

• GNU scientific libraries (gsl, gslcblas) for random number generation and histogram
• HDF5 (hdf5, hdf5_hl, H5Cpp) for saving observables to binary file formats
• Boost for python wrapping, logging and stuff (Boost.python, Boost.log, Boost.system, Boost.filesystem, Boost.multi_array)

Requirements for usage:

• Python
• h5py

Building

The way to build this using cmake goes like

$ mkdir build
$ cd build
$ cmake .. $CMAKE_FLAGS
$ make -j
$ cd ..

Any user specific configuration goes into the $CMAKE_FLAGS and should be defined beforehand.

Usage

Import revreaddy in a python script, e.g.

>>> import revreaddy 
>>> s = revreaddy.Sim()
>>> s.run(steps=10000, timestep=0.1)