The AutoReduce Library

This chapter documents the public modules, classes, and functions that make up AutoReduce.

Systems

Core symbolic representation of nonlinear dynamical systems.

System

Class that stores the system model in this form: x_dot = f(x, theta), y = Cx.

Conversion from python-control input/output systems.

from_nonlinear_io_system

Create an AutoReduce System from a NonlinearIOSystem.

Solvers

Numerical solvers for ordinary differential equation systems.

ODE

To solve the Model using scipy.odeint

Local sensitivity analysis solvers.

SSM

Class that computes local sensitivity analysis coefficients for the given Model using a numerical approximation method discussed in doi: https://doi.org/10.1016/0021-9991(76)90007-3 Uses numerical method to find sensitivity analysis matrix (SSM).

Reductions

Time-scale separation and model-reduction orchestration.

`Reduce`	The class can be used to compute the various possible reduced models for the System object and then find out the best reduced model choice using doi : https://doi.org/10.1101/640276
`ReduceUtils`	For various utility methods developed on top of Reduce class and other utility functions
`solve_timescale_separation`	Solve time-scale separation directly from a `System`.
`sympy_variables_exist`	Check whether variables appear in symbolic ODE expressions.
`sympy_solve_and_substitute`	A function to solve using sympy and substitute into the equation
`sympy_get_steady_state_solutions`	Solve for each collapsed_variable from corresponding collapsed_dynamics one by one.
`create_system`	Create a base `System` from symbolic model data.

Conservation-law reduction methods.

`find_conserved_sets`	Find conserved species sets by summing ODE terms.
`setup_conservation_laws`	Create conservation-law expressions from conserved species sets.
`apply_conservation_laws`	Apply conservation laws to a system in place.
`solve_conservation_laws`	Find and apply conservation laws to a `System`.

Abundance-based model-reduction methods.

solve_approximations

Raise until abundance-based reduction is implemented.

Projection-based reduced models using PyDMD.

`DMDReduction`	Fitted PyDMD model and the data used to construct it.
`fit_dmd`	Fit a dynamic mode decomposition model to state snapshots.
`fit_dmdc`	Fit a dynamic mode decomposition with control model.

Utilities

Model import and symbolic conversion utilities.

`load_ODE_model`	Directly load ODE with sympy
`ode_to_sympy`	Returns Sympy object for the given ODE function
`sympy_to_sbml`	Return an SBML document for a SymPy-backed model.
`load_sbml`	A function that takes in an SBML file and returns x,f,P,params_values.

Convenience constructors for solvers and reducible systems.

`get_ODE`	For the given timepoints, create an ODE class object for this System object.
`solve_ODE_SSM`	For the given timepoints, returns the full solution (states, sensitivity coefficients, outputs)
`get_SSM`	For the given timepoints, create an SSM class object for this System object.
`get_reducible`	Create a `Reduce` object from a base `System`.
`solve_ode`	Solve a `System` over the given timepoints.
`solve_ssm`	Compute the local sensitivity matrix for a `System`.
`solve_sensitivity`	Solve the coupled state and local sensitivity equations.
`reduce_utils`	Create a `ReduceUtils` object from a `Reduce` object.
`get_ode_solutions`	Solve a system over time and return transposed state trajectories.

SBML model construction utilities.

`create_sbml_model`	Creates an SBML Level 3 Version 2 model with some fixed standard settings.
`add_species`	Add species to SBML model
`add_reaction`	Add reaction to given model
`add_parameters`	Adds global parameters to the SBML Model