The T-Rex software and web service (Makarenkov 2001) implement a number of phylogenetic methods including a special technique for constructing reticulograms. The latter is commonly referred to as the T-Rex method.
The T-Rex method (Legendre and Makarenkov 2002) begins by building a phylogenetic tree using an existing tree reconstruction technique, such as Neighbour joining (NJ). Additional reticulation edges are then added to the tree in order to achieve a better representation of the taxon–taxon pairwise distances computed from the data. For example, after building a NJ tree, a pair of taxa, say A and B, may be placed in the tree in a way that exaggerates their pairwise distance. To rectify this, one can add an edge between A and B, or nodes adjacent to them, depending on which choice leads to the best overall match between the distances represented by the tree and distances computed from the data. The number of reticulation edges to be added can be given by the user or determined automatically using various different criteria also implemented in the T-Rex software package.
– Makarenkov, Vladimir. 2001. “T-Rex: Reconstructing and Visualizing Phylogenetic Trees and Reticulation Networks.” Bioinformatics 17: 664–668.
– Legendre, Pierre, and Vladimir Makarenkov. 2002. “Reconstruction of Biogeographic and Evolutionary Networks Using Reticulograms.” Systematic Biology 51 (2): 199–216.