Skip to main content
Fig. 4 | Algorithms for Molecular Biology

Fig. 4

From: Statistically consistent divide-and-conquer pipelines for phylogeny estimation using NJMerge

Fig. 4

Divide-and-conquer pipeline using NJMerge. We present a divide-and-conquer pipeline that operates by (1) estimating distances between pairs of species using method \(M_D\), (2) decomposing the species set into pairwise disjoint subsets, (3) building a tree on each subset using method \(M_T\), and (4) merging trees together using the distance matrix using NJMerge. Step 2 can be performed by estimating a tree from the distance matrix (e.g., using NJ) and then decomposing this tree into pairwise disjoint subsets of species (shown in blue). Although not explored in this study, this pipeline can be run in an iterative fashion by using the tree produced in Step 4 to define the next subset decomposition. In this schematic, sets of species are represented by circles, distance matrices are represented by squares, and trees are represented by triangles

Back to article page