This means that the distance matrices generate an association scheme.

Based on a distance matrix relating the taxa, calculate as follows:

We then proceed to updating the distance matrix, by computing according to the above formula for every node .

The resulting distance matrix may then better fit an alternate (presumably less optimal) tree.

Often the distance matrix of size is materialized to avoid distance recomputations.

Initially, a distance matrix containing all the pairwise distances between the genes is calculated.

After each step, a new distance matrix between the newly formed clusters and the other clusters is recalculated.

The tree nodes are then replaced by two and the distance matrix reduced.

Like neighbor joining, the method takes a distance matrix as input, and works by agglomerating clusters.

The distance (or cost) matrix contains information about distances of the edges.