Local Search by Concentration Controlled DNA Computing

Concentration controlled DNA computing are presented for realizing a local search for thesolution of a shortest path problem.In this method, the concentrations of DNAs representingedges are determined according to the costs on the edges, and then the hybridization processis performed.Since the concentrations of hopeless candidate solutions tend to be small inthe result of the hybridization process, we can realize a local search by the concentrationcontrolled DNA computing.In order to discuss about the relation between given costs onedges in the graph and concentrations of generated DNA paths, the simulation model of thehybridization process is used, and the result of laboratory experiment is shown.