SOSlib 1.6.0 :: The Tokyo Release Series
(December 17, 2005,
Vienna/Oxford/Leipzig/Linz)
SOSlib is both a programming library (API) and a set of command-line applications for symbolic and numerical analysis of a system of ordinary differential equations (ODEs), derived from a (bio)chemical reaction network encoded in SBML. This release newly features basic sensitivity analysis routines! Detailed documentation for installation on various platforms, for application usage as well as as for the API is available at our new website at http://www.tbi.univie.ac.at/software/odeSolver/. It is written in ISO/ANSI C, can be build with GNU tools (autoconf/automake) but also comes with Visual C++ project files. Windows binary executables are available. The API allows full access to all internal structures and routines. Bindings for scripting languages are in development.
SOSlib extends libSBML functionality to providing API functions
that: A fine-grained wrapper around the well established SUNDIALS library provides fast and reliable numerical integration of stiff and non-stiff ODE systems with `BDF' or `Adams-Moulton' methods to solve for x(t). Sensitivity analysis is provided via SUNDIALS implementations of `staggered' or `simultaneous corrector' methods to calculate forward sensitivities of a given variable's integrated solution to parameter values (S=δx(t)/δp). Adjoint sensitivity gradients and handling of DAE systems (SBML algebraic rules) are in development. SOSlib is an open source project hosted by sourceforge and distributed under the LGPL license. It provides researchers and application developers with easy-to-use and highly efficient interfaces to well-established methods of theoretical chemistry/biology and systems theory to enable rapid scientific progress in the new field of Systems Biology. Developers and users are warmly welcome to contribute and/or criticize and/or suggest priorities for further development! SOSlib is already used for high-throughput integration of evolutionary models. It is programmed for use in analytic applications (such as parameter optimization or feedback and bifurcation analysis) and within hybrid or multi-scale solvers or merely as an efficient back-end for systems biological GUI tools, such as already implemented for the CellDesigner. The actual usage of the library has been and will be strongly influenced by user feedback and suggestions! Please don't hesitate to contact us via one of the possibilities given at http://www.tbi.univie.ac.at/software/odeSolver/. Rainer Machné |