Triplet Supertrees by

Phylogenetic trees are used in the interdisciplinary field of bioinformatics to relate and classify species according to evolutionary history. Constructing large trees is infeasible because the search space on trees is very large. As a result, supertree methods have been the focus of much research. By building trees from smaller, already constructed trees, supertrees are a feasible way of constructing large trees. Several supertree methods have previously been developed. One, called Q(I)LI, decomposes its input trees to quartet trees, which are trees of four species. The supertree is assembled by combining all the quartet trees. This thesis reports on the design and implementation of a new algorithm called T(I)LI which uses triplets instead of quartets. In an experimental study, it is shown that T(I)LI outperforms Q(I)LI in efficiency, and under som