Parallelisation of a multi-neighbourhood local search heuristic for a phylogeny problem.

In this work we study a phylogeny problem. That is, given a collection of organisms, we want to reconstruct the evolutionary history of the organisms. We are interested in inferring relationships between the organisms. For a number of reasonable biological hypotheses the problem becomes NP-hard. Besides that, the problem data is large enough to inhibit anyone using exact algorithms to solve, in practical computational time, real instances of the problem. In this work, we propose an innovative technique based on local search procedures that use multiple starts and diversified neighbourhoods.