Skip to main content
Figure 1 | EURASIP Journal on Bioinformatics and Systems Biology

Figure 1

From: A comparison study of optimal and suboptimal intervention policies for gene regulatory networks in the presence of uncertainty

Figure 1

Optimization procedure for an OBR policy. We start with a hyperstate (z n ,α n ). We calculate P ̄ using α n and utilize the successive approximation method for a fixed K to find an optimal action a n ∗ . We then apply the action a n ∗ to the network and let it transition from state z n , or z n ~ depending on the optimal action, to a new random state zn+1 according to P ̂ . We incorporate the new observation into our prior knowledge and update the hyperparameter matrix to αn+1 by incrementing the entry at (z n ,zn+1) or ( z ~ n , z n + 1 ) of the hyperparameter matrix α n by 1. We repeat the entire optimization procedure, but now with the new hyperstate (zn+1,αn+1).

Back to article page