SimBiology® 2.3

Product Description

• Introduction and Key Features
• Working with SimBiology®
• Building a Model
• Running a Simulation
• Analyzing a Model
• Saving and Exporting Results

Running a Simulation

Together, SimBiology and MATLAB provide a wide range of solvers for simulating stochastic and deterministic systems.

Stochastic Solvers

SimBiology provides three types of stochastic solvers: a stochastic simulation algorithm (SSA), explicit tau-leaping, and implicit tau-leaping.

SSA simulates one reaction at a time based on the propensity function for each reaction. It is based on the Gillespie exact stochastic simulation algorithm.

Explicit tau-leaping automatically chooses the time interval tau, so that the relative change in the propensity function is less than a user-defined error tolerance. Often used for solving large, numerically nonstiff systems, explicit tau-leaping provides faster simulation speed than SSA, but represents an approximation to a true realization of the system dynamics.

Implicit tau-leaping is similar to explicit tau-leaping but works optimally with numerically stiff systems. When solving numerically stiff systems, this solver remains stable at larger time intervals than does explicit tau-leaping.

Deterministic Solvers

SimBiology provides solvers for stiff and nonstiff systems. In addition to the traditional MATLAB solvers, SimBiology adds the CVODE solver from the SUNDIALS suite of solvers.

Stiff solvers include numerical differentiation formulas (NDFs), a modified Rosenbrock formula of order 2, an implementation of the trapezoidal rule using a free interpolant, and an implementation of TR-BDF2 (an implicit Runge-Kutta formula with a first stage that is a trapezoidal rule step and a second stage that is a backward differentiation formula of order 2).

Nonstiff solvers include Dormand-Prince, Bogacki-Shampine, and Adams-Bashforth-Moulton.

Simulation Events

Simulation events let you define a sudden change in model behavior. You can specify that an event occurs at a specific simulation time or in response to a change within the system, such as the value of a species.

Ensemble Runs

Ensemble runs let you analyze model output over multiple simulations—helpful when you are dealing with stochastic simulations or need to perform parameter scans on your model. You specify the model and number of iterations as inputs. You can calculate the mean and variance of the ensemble data, and visualize your ensemble results with distribution and mountain plots.

Simulation Settings

You can specify the solver type, simulation start and stop times, and other simulation options. You can also specify whether dimensional analysis should be performed on the model as a verification step. Dimensional analysis checks that all mathematical expressions in the model are in consistent dimensions and then performs unit conversions automatically.

Next