MixTreEM: an acronym for Mixture of Trees using Expectation Maximization, is a species tree reconstruction method.
It uses a probabilistic generative (mixture) model consisting of two phases. In the first phase, a set of probable species trees are inferred given gene family data. In the second phase, each of the species tree, along with the gene families, is fed into DLRS model (implemented as primeDLRS in PrIME and Delerious in JPrIME), ultimately giving us the best species tree. For detailed discussion of model and algorithms, you may read the paper.
MixTreEM is implemented in an MPI-based parallel architecture using Boost C++ libraries. Both the source and the executables are open source which means you may download them free of cost.