SPECIES TREE INFERENCE FROM GENOMIC SEQUENCES USING THE LOG-DET DISTANCE.

Area of Science:

• Phylogenetics
• Computational Biology
• Genomics

Background:

• The log-det distance was previously established for gene tree inference under simple models.
• Inferring species trees from genomic data is complex due to factors like incomplete lineage sorting and varying evolutionary processes.

Purpose of the Study:

• To demonstrate the efficacy of the log-det distance for species tree inference under complex mixture models.
• To validate the log-det distance as a robust tool for analyzing large-scale genomic sequence data.

Main Methods:

• Utilized the log-det distance metric.
• Employed distance-based tree construction algorithms.
• Applied methods to aligned genomic-scale sequence data, including multiple genetic loci.

Main Results:

• The log-det distance, combined with distance-based methods, enables consistent species tree inference.
• The method is effective even when gene trees differ due to incomplete lineage sorting and varied substitution processes.
• Demonstrated applicability to ultrametric species trees.

Conclusions:

• Log-det distance-based methods provide a statistically consistent approach for species tree inference.
• The speed and simplicity of these methods make them valuable benchmarks for more complex phylogenetic inference techniques.