NJst-J

NJst-J infers species trees from collections of gene trees by incorporating nonbinary unrooted constraint trees to mitigate gene tree discordance under the multispecies coalescent model.

Key Features:

• Integration of External Constraints: Incorporates nonbinary unrooted constraint trees to include partial prior knowledge of species-tree structure during estimation.
• Statistical Consistency: Is statistically consistent under the multispecies coalescent model when subject to external constraints.
• Computational Efficiency: Achieves substantially faster runtimes than methods such as ASTRAL-J, enabling analysis of large datasets.
• Accuracy: Produces species-tree estimates with accuracy comparable to or exceeding that of ASTRAL-J in comparative analyses.

Scientific Applications:

• Large-scale phylogenomics: Applies to projects with many species and genes, such as the Avian Phylogenomics Project, by enabling rapid species-tree inference from discordant gene trees and reducing runtimes from hours to minutes or seconds.

Methodology:

Processes nonbinary unrooted constraint trees to incorporate external constraints and performs species-tree estimation consistent with the multispecies coalescent model.