Setup

Clone the LpAdaptation repository

git clone https://github.com/breezerider/LpAdaptation_Code.git ~/LpAdaptation_Code

Follow these steps to build pSSAlib compatible with MATLAB. Clone the pSSAlib repository to ~/pSSAlib

git clone https://github.com/breezerider/pSSAlib.git ~/pSSAlib

and run:

cd ~/pSSAlib
./configure --without-libsbml --prefix=~/.local/ --disable-cli --with-pic
make clean; make -j && make install

Data for the Figures

All the data is available in the data subdirectory.

Reproduce Data for Figure 1

To reproduce Figure 1, please change directory to figure-1 and run

make

which will produce a shell script run_fig1.sh. Run this cripts and it will produce the true-negative and true-positive counts, that can be plotted using plot.py.

Reproduce Data for Figure 2

You can reproduce the results in Figure 2 by changing your working directory to figure-2/{a,b,c} and running

make compile

in the respective subdirectories and running the respective scripts in the work subdirectory. Makefiles allow one to submit individual parameter estimation runs as SLURM jobs by running:

make run

However, these MATLAB scripts can also be run interactively. First, please make sure that LpAdaptation_Code directory is on the MATLAB path. Then in MATLAB just run the .m scripts from the respective work subdirectory. Some paramaters that you ay want to change:

• nSampleID -- sample identifier
• strMethod -- method name, 'dc' or 'cmaes'
• strOutputPath -- path to store the output
• strOutputName -- output file name prefix
• dTimeStart -- first time point included in the steady-state trajectory
• dTimeStep -- time step
• dTimeEnd -- last time point included in the steady-state trajectory

Extensions

New models can be added to models.h and respective MATLAB driver script, following the examples presented in this repository.