A Particle Swarm Optimizer for Finding Minimum Free Energy RNA Secondary Structures

This paper introduces the HelixPSO particle swarm optimization (PSO) algorithm for finding minimum energy RNA secondary structures. It is shown experimentally that HelixPSO profits when it is combined with a genetic algorithm that finds a good starting population for HelixPSO. On all test instances this hybrid variant of HelixPSO performs significantly better than a state-of-the-art genetic algorithm. Also compared with another PSO algorithm that has been proposed very recently for the prediction of RNA secondary structures, HelixPSO is more efficient both in terms of free energy and correctly predicted base pairs