AleRax

AleRax performs joint estimation of gene trees and a shared species tree using a probabilistic model that accounts for gene duplication, horizontal gene transfer, and gene loss while integrating sequence-substitution information and a joint phylogenetic-reconciliation likelihood.

Key Features:

• Probabilistic Modeling: Employs a probabilistic framework that models gene duplications, transfers, and losses for reconciliation of gene trees with a species tree.
• Uncertainty Management: Accounts for uncertainty in gene trees derived from short sequences via an efficient approximation that computes the joint phylogenetic-reconciliation likelihood and enables sampling of reconciled gene trees.
• Performance Efficiency: Achieves computational performance approximately one order of magnitude faster than competing gene tree inference tools while maintaining comparable accuracy.
• Robustness Under Uncertainty: Exhibits superior robustness compared to SpeciesRax and ASTRAL-Pro 2 under conditions of gene-tree uncertainty.
• Parallel Processing Capability: Supports parallel processing of multiple gene families, simultaneous comparison of competing phylogenetic hypotheses, and estimation of duplication, transfer, and loss parameters across genome-scale datasets with hundreds of taxa.

Scientific Applications:

• Species phylogeny inference: Co-estimates species trees together with reconciled gene trees to infer evolutionary relationships among species.
• Genome evolution analysis: Investigates mechanisms of genome evolution, including gene duplication, horizontal gene transfer, and gene loss.
• Comparative genomics: Facilitates comparative analyses across diverse taxa by reconciling gene trees with species trees to test evolutionary hypotheses.

Methodology:

Uses a probabilistic duplication-transfer-loss (DTL) model to simultaneously infer reconciled gene trees and a shared species tree, employs an efficient approximation to compute the joint phylogenetic-reconciliation likelihood and sample reconciled gene trees, and supports parallel processing of multiple gene families.