Scripts to perform dynamic optimization of a model of invasive aspergillosis

Matlab scripts to perform optimization and analyses of a dynamic optimization model of invasive aspergillosis:
"Dynamic optimization reveals alveolar epithelial cells as key mediators of host defense in invasive aspergillosis"
Jan Ewald, Flora Rivieccio, Lukáš Radosa, Stefan Schuster, Axel A. Brakhage, Christoph Kaleta
https://doi.org/10.1101/2021.05.12.443764

File structure:

--- Core optimization files ---
-- simulation.cpp --
Description of ODE model in 'C', used by solver to calculate the time course given the parameters, initial states and control variables

-- simulation.mexa64 --
Compiled mex-File from simulation.cpp used as interface by MATLAB to 'C' files.

-- optFun.m --
Definition of the objective function. Uses a control vector 'u' to calculate the, objective function value F, the time course of state variables 'x' and gradients used by the solver to determine the optimal control of 'u'.

-- single_opt.m --
Function which performs an optimization run for a given parameter set and a randomized initial solution u0. Optimization can be conducted by SNOPT or in conjunction with IPOPT. Both solver are not provided here (proprietary).

--- Additional files ---
-- solve_with_ipopt.m --
Configuration file and interface to call the solver IPOPT.

-- lognorm_mu.m --
Function to provide log-normal distributed samples given the standard deviation, reference value (mode of distribution) and number of samples which should be calculated.

-- mice_param_alv.m --
Reference parameter set of the model for mice as host organism and cell numbers normalized per alveolus.

-- parameter_model.m --
Definition of parameter container to handle interface to optFun-call by solvers which do not allow parameter in function call.

--- Optimization Runs ---
-- rand_params.m --
Performs optimization runs for multiple randomized parameter sets and several scenarios to determine parameter sensitivity.

-- doseresponse.m --
Performs optimization runs for a sequnce of initial conidial dosages to determine the dose response of the system.

--- Scripts for post-analysis ---
-- rand_analysis.m --
Loads the data created by the optimization runs (rand_params.m) and performs parameter sensitivity analysis as well as produces plots for analysis related to parameter influence.

-- doseanalysis.m --
Loads the data created by the optimization runs (doseresponse.m ) and produces dose response curves as well as additional plots and analyses.

--- Optimization output files ---
-- rand_param_out/ --
Folder contains optimization results as MATLAB session-file for randomized parameters used for parameter sensitivity analysis.

-- dose_out/ --
Folder contains optimization results as MATLAB session-file for dose response curve calculation.