Examples Overview
examples_overview.RmdWe here provide some examples to simulate, calibrate or analyse different models. The calibration (parameter estimation) process includes uncertainty quantification (UQ) and it can be followed by global sensitivity analysis (SA). You find more details about the different tools in the accompanying article.
The examples are listed in order of increasing difficulty and run time.
AKAR4
These examples can be run on a laptop.
- Simulate AKAR4 deterministically Builds the AKAR4 reaction network as an ODE and simulates it with an ODE solver.
- Simulate AKAR4 stochastically Builds the AKAR4 reaction network as a stochastic model and simulates it with the Gillespie algorithm.
- UQ of deterministic AKAR4 AKAR4 is calibrated using Markov chain Monte Carlo (MCMC) methods in a likelihood-based approach.
- UQ of stochastic AKAR4 AKAR4 is calibrated with an Approximate Bayesian Computation (ABC) approach.
- SA of AKAR4 Global sensitivity analysis is performed on the deterministic AKAR4 model using independent prior parameter distributions.
- UQ and SA of AKAR4 First UQ and then SA is performed on the deterministic AKAR4 model.
AKAP79
These examples are meant to be run on a compute cluster, but with some patience you can test them on a multicore workstation or laptop.
- Parallel chains with MPI Describes how to run the package in parallel.
- UQ and SA of AKAP79 deterministic model
- UQ and SA of AKAP79 stochastic model
The Documentation article Parallel chains with
MPI includes a script that needs to be present as a file to
be able to execute it with mpirun, so the order of these
articles is important.
CAMKII
The demo R scrips can be found in the folder inst/extdata.
The ?uqsa::uqsa_example function can locate the example
folders after package installation.