PhytClust: Efficient and Optimal Monophyletic Partitioning of Rooted Phylogenetic Trees.

PhytClust partitions rooted phylogenetic trees into monophyletic clusters by minimizing within-cluster dispersion, achieving exact global optima for a fixed cluster count and scaling to over 100,000 taxa.

• Why it matters: Many existing tree-clustering methods depend on user-chosen distance thresholds or heuristics, making results hard to justify and inconsistent across datasets.
• What they did: PhytClust is threshold-free: for a specified number of clusters it computes the exact global optimum for the low-dispersion objective, then picks the cluster number using a cluster-validity index.
• The result: In simulations, it outperforms existing methods in accuracy, stays competitive in speed, and produces reproducible partitions that reflect both topology and branch lengths.

AI summary of the abstract · may contain errors · not medical advice · read the paper.