This is the repository containing part of the MATLAB code used for PhD Thesis of @SaraNapolitano (from @dibbelab). The other part (regarding the cybergenetics controller to synchronise the cell-cycle) is available at the repository Cycloop
.
It requires only a standard computer with enough RAM to run the code in the MATLAB environment.
- MATLAB R2019a
- Additional MATLAB toolboxes (e.g. Signal Processing Toolbox, Image Processing Toolbox, etc.)
- Download the repository from https://github.com/dibbelab/s.napolitanoPhd.git .
- Unpack the files.
- Start MATLAB and navigate to the
TFEB_QuantitativeAnalysis
folder.
Set the working directory to ./TFEB_QuantitativeAnalysis/Fitting/
.
- Run the script
Main_Fitting.m
to derive the values of the time constant used for mTOR inhibition model and reported in Table 4.1. Moreover it generates also the simulations shown in Fig. 4.14B and Fig. 4.14C, i.e. the simulation with the biological hypothesis on the synergistic effect.
Set the working directory to ./TFEB_QuantitativeAnalysis/Simulator/
.
- Run the script
Main_Simulator.m
to generate the simulations shown in Fig. 4.6B and Fig. 4.7B, i.e. the simulation without and with the feedback hypothesis, and the simulations shown in Fig 4.15 A-B, i.e. the simulation of TFEB nuclear translocation when the cells are treated with Torin1 in combination with CHX or Bort.
The folder ./TFEB_QuantitativeAnalysis/ExperimentsAnalysis/Segmentation/
contains the scripts used for segmentation of human HeLa cells. Note that it is needed to run FastER (https://bsse.ethz.ch/csd/software/faster.html) segmentation before the MATLAB script.
The folder ./TFEB_QuantitativeAnalysis/ExperimentsAnalysis/MakeFigure/
contains the scripts used to clear the data-set and to generate the images with experimental data shown in the Chapter 4 of the PhD Thesis