ByoDyn: integrative tool for Systems Biology

In systems biology it is becoming a routine task to build models of increasing complexity on a given biochemical network or pathway of interest. One of the main problems in building such models is the determination of the parameters underlying each modelled process. ByoDyn has been designed to provide an easily extendable computational framework to estimate and analyze parameters in highly uncharacterized models.

The current version 5.12 of ByoDyn includes a set of tools to 1) integrate ordinary differential equations (ODEs), including systems with events, rules (differential algebraic equations, DAE) and delays built from a given biological model; 2) stochastic simulators: SSA and tau-leap; 3) globally optimize the parameters that fit the provided experimental information and evaluate the sensitivity of the model with respect to the different parameters; 4) include the sensitivity of the parameters in an optimal experimental design pipeline based on the Fisher information matrix; and 5) Monte Carlo sampling coupled with cluster analysis and PCA to determine the global shape of the parameter landscape. The program makes use of external software, providing a Python binding schema that allows the user to easily implement new software in the desired calculation protocol. The program benefits from its interface with the SBML library, which ensures communication with other existing tools in the field.

ByoDyn (released under the GPL) is a Python based program built on top of several open source software, PORT (FORTRAN library from Netlib), SciPy and libSBML, matplotlib, ScientificPython and it provides easy binders to OpenModelica, XPPAUT, Octave and Gnuplot. The program has been tested in Linux Fedora and in Mac OS X platforms, having been parallelized in a Sun Grid Engine environment and on a OpenMPI standards. ByoDyn can be run on-line by accessing http://byodyn.imim.es.

Please follow this link to ByoDyn project page at SourceForge.net