Application of different multi-objective decision making techniques in the phylogenetic inference problem

The phylogenetic inference aims to estimate the evolutionary relationships among different species. These relationships are commonly represented as a phylogenetic tree. The phylogenetic inference has been treated in bioinformatics as an optimisation problem. Different criteria have been proposed to define the optimal tree between the possible topologies. In order to reduce the bias associated to the dependency of a specific criterion, different multi-objective optimisation strategies have been proposed. However, they result in multiple solutions which represent a trade-off between conflicting criteria. It is a problem in the biological context, because the selection of one representative solution can result biased when an arbitrary criterion is used. Multi-objective decision making techniques have been successfully used to deal with this problem in other areas of research, but their application in phylogeny has not been studied. In this work, we adapt three multi-objective decision making techniques to deal with the multi-objective phylogenetic inference problem. Their evaluation is performed considering the objective and the solution space, known as tree-space. The results show that the multi-objective decision making techniques applied over the objective space do not consider the topological features of the trees. This result uncovers the need for the designing of new strategies which consider information from the tree-space in the evolution of the algorithms.