A new efficient algorithm for inferring explicit hybridization networks following the Neighbor-Joining principle

Area of Science:

• Evolutionary biology
• Bioinformatics
• Computational phylogenetics

Background:

• Traditional phylogenetic trees cannot fully represent complex evolutionary events like hybridization, recombination, and horizontal gene transfer.
• Phylogenetic networks are necessary to accurately model these non-tree-like evolutionary mechanisms.
• Existing methods may not efficiently or accurately infer these complex evolutionary histories.

Purpose of the Study:

• To develop a novel, efficient heuristic algorithm for inferring hybridization networks from evolutionary distance matrices.
• To improve the representation of complex evolutionary processes in phylogenetic analysis.
• To provide a method that is exact for tree metrics and effective for complex scenarios.

Main Methods:

• The algorithm integrates the Neighbor-Joining concept with a least-squares criterion for network construction.
• It incorporates a novel step to detect potential hybridization events before joining nodes.
• The method was tested on simulated and real biological datasets, including large trees (32 and 64 leaves).

Main Results:

• The algorithm correctly identifies the exact tree solution when the input data represents a tree metric.
• It demonstrates high accuracy in recovering hybridization events for large datasets, using both distance and sequence data.
• Simulations and real-world data analyses show the algorithm's effectiveness and robustness.

Conclusions:

• The proposed heuristic algorithm offers an efficient and accurate approach to inferring hybridization networks.
• This method advances the modeling of complex evolutionary histories beyond traditional phylogenetic trees.
• The algorithm provides a valuable tool for understanding evolutionary mechanisms involving hybridization.