This library is developed to produce the results for the paper,
The code has been tested successfully with versions of Python (at least 2.7) on Windows and macOS.
The subfolders contains codes to generate the results obtained by numerical simulations in the main and supplementary figures.
Reaction Diffusion 1D : Kymographs in Fig. 1g, 8b and c, Supplementary Fig. 1h, i, 9d-g
MT Monte Carlo : Supplementary Fig. 1g
Aurora Reaction Diffusion 2D_2 variable + Recurrence Quantifications : Fig. 3i, Supplementary Fig. 4h. The spatial recurrence quantification code is used for the quantification shown in Fig. 3f
Aurora Reaction Diffusion 2D_3 variable : Supplementary Fig. 4g
Each folder has a file with name ending '_main.py'. This file combines different objects (functions) defined in linked files in the same folder. For example, in the folder SynMMS/Reaction Diffusion 1D, the 'reaction_diffusion1D_main.py' loads the model equations and parameters from 'model.py' file. Sample code :
pbc = PeriodicBoundaryConditions()
model = SynMMS()
light = TimeVaryingCue()
rd = ReactionDiffusion1D(model, pbc, light)
rd.simulate(save=True, animate=None)One can either animate/save or do both simultaneosly. In the case when 'save=True', a subfolder needs to be created to save the files in '.npy' format. 'plot_kymograph.py' can be used to generate the kymographs from 'saved data' folder.
The subfolder 'Aurora Reaction Diffusion 2D_2 variable + Recurrence Quantifications' has two main files. 'reaction_diffusion2D_main.py' to genrate the 2D pattern. 'spatial_recurrence_plot_main.py' to caculate Information Entropy value corresponding to an already saved spatial pattern.
Excecuting codes via command prompt (eg. Anaconda command prompt) is recommended.
The folder "CellularAutomata_RDSim" contains Matlab R2018a scripts and c++ code for compilation in Matlab to simulate the reaction diffusion system of the tubulin-stathmin association/dissociation cycle and the stathmin phosphorylation cycle in 1D and 2D (Fig 4f, Supplementary Fig. 6i-k and 8i). This code has been successfully tested with Matlab R2018a - R2020a under Mac OS X and Windows 10. It requires a c++ compiler (xCode for Mac OS X, Visual Studio 2019 for Windows) to be configured to compile and execute c++ code from within Matlab.
Please refer to the respective Matlab scripts on how to use the simulations to generate figure panels.