Skip to main content


Fig. 1 | Biology Direct

Fig. 1

From: Long range personalized cancer treatment strategies incorporating evolutionary dynamics

Fig. 1

Example of single-step heuristics, multi-step heuristics, and global optimization on a decision tree. The example tree is spanned by 4-step treatment sequences with only two possible dosage combinations (full drug 1 dosage (1, 0) and full drug 2 dosage (0, 1)) in each step. Each node represents a population structure and the two edges emitting from each internal node represent the two possible treatments following the observed population structure. A path represents a treatment sequence. The treatment sequences traversed by three strategies are marked with distinct colors: green – one-step heuristic, red – two-step heuristic, blue – global optimization algorithm. The paths with bold colored lines are the treatment sequences selected by the strategies. A one-step heuristic chooses one of the two branching edges in each time step. A two-step heuristic chooses one of the four branching paths of length 2 in every two steps. Global optimization keeps the two best paths of length 2 at step 2, tracks the subsequent 4 paths for each earlier path and chooses the optimal one

Back to article page