Protocol: discrete dynamic model
Created by Ming Wu, 07/2012
First one need to have a signaling pathway map in which the activation/inhibition relationship should be clearly defined. Then translate the map into anadjacency matrix, for example:
element (i,j) is 1 if i activates j, -1 if i inhibits j, 0 otherwise.
The matrix can then be input into MATLAB.
Our code for simulating a discrete dynamic model is:
To run the code, one can use:
The initial conditions should be specified, 1 for activated, 0 for control, -1 for inhibited, 9 for unknown (will be randomly generated)
The "rule" matrix is the adjacency matrix abovementioned.
one need to also define the population (No. of different simulations: 500 different cells), the number of steps to compute, and if want to plot some of the components.
In many cases the rule matrix is sparse, thus I have another code to generate the matrix by giving only total number of the components, and a three column vector (i,j,k) meaning component i regulate component j in the form of k (+1/-1).