Difference between revisions of "Simulation"
From BioUML platform
Ilya Kiselev (Talk | contribs) (Created page with "Currently, BioUMl contains next simulationa engines: *ODE Simulation Engine *Stochastic Simulation Engine *Agent-based Simulation Engine") |
Ilya Kiselev (Talk | contribs) |
||
| Line 1: | Line 1: | ||
| − | Currently, BioUMl contains next | + | Currently, BioUMl contains next simulation engines: |
| − | * | + | *ODE Simulation Engine |
| − | * | + | *Stochastic Simulation Engine |
| − | *[[Agent-based | + | *Agent-based Simulation Engine |
| + | |||
| + | == ODE Simulation Engine == | ||
| + | Generates ODE system with descrete events. Along with the initial model state it constitutes Initial Value Problem (IVP) which can be solved with appropriate methods implemented in BioUML: | ||
| + | *Euler | ||
| + | *Dormand-Prince | ||
| + | *RADAU IV | ||
| + | *JVODE | ||
| + | |||
| + | It is also applicable for [[modular models]] created in BioUML and satisfying certain limitations on modules. In that case [[flattening algorithm]] is applied before simulation. | ||
| + | |||
| + | == Stochastic Simulation Engine == | ||
| + | Interpretes chemical reactions in the model as stochastic processes. Generated code implements interface for stochastic solvers: | ||
| + | *Simple Gillespie | ||
| + | *Gibson-Bruck | ||
| + | *Tau-leaping | ||
| + | *Maximum time-step | ||
| + | |||
| + | == Agent-based Simualtion Engine == | ||
| + | Is applicable for [[modular models]] regardless modules implementation. Each module is simulated with its own simulation engine, scheduler provides their intercommunications. | ||
Revision as of 15:31, 26 March 2013
Currently, BioUMl contains next simulation engines:
- ODE Simulation Engine
- Stochastic Simulation Engine
- Agent-based Simulation Engine
ODE Simulation Engine
Generates ODE system with descrete events. Along with the initial model state it constitutes Initial Value Problem (IVP) which can be solved with appropriate methods implemented in BioUML:
- Euler
- Dormand-Prince
- RADAU IV
- JVODE
It is also applicable for modular models created in BioUML and satisfying certain limitations on modules. In that case flattening algorithm is applied before simulation.
Stochastic Simulation Engine
Interpretes chemical reactions in the model as stochastic processes. Generated code implements interface for stochastic solvers:
- Simple Gillespie
- Gibson-Bruck
- Tau-leaping
- Maximum time-step
Agent-based Simualtion Engine
Is applicable for modular models regardless modules implementation. Each module is simulated with its own simulation engine, scheduler provides their intercommunications.
