Delimiting Coalescence Genes (C-Genes) in Phylogenomic Data Sets

Area of Science:

• Genomics
• Phylogenetics
• Evolutionary Biology

Background:

• Coalescence methods are popular for inferring species trees from large genomic datasets.
• These methods account for incomplete lineage sorting (ILS), a key factor in evolutionary history.
• A critical assumption is free recombination among, but not within, coalescence genes (c-genes).

Purpose of the Study:

• To evaluate the impact of c-gene length and recombination on the statistical consistency of summary coalescence methods.
• To investigate the 'recombination ratchet' phenomenon in species tree inference.
• To determine the appropriateness of using complete protein-coding sequences as c-genes.

Main Methods:

• Analysis of theoretical assumptions underlying summary coalescence methods.
• Review of empirical estimates of recombination breakpoints.
• Examination of the relationship between c-gene size, taxon sampling, and phylogenetic accuracy.

Main Results:

• Statistical consistency of summary coalescence methods is not guaranteed if recombination assumptions are violated.
• Longer loci, such as complete protein-coding sequences, may not be valid c-genes due to intralocus recombination.
• The 'recombination ratchet' suggests shorter c-genes are necessary with increased taxon sampling.

Conclusions:

• Complete protein-coding sequences are often inappropriate as c-genes for summary coalescence methods, especially in species-rich clades.
• Violation of the no-recombination-within-loci assumption can decrease the probability of accurate species tree inference.
• Recombination is essential for generating the independent gene histories needed for robust species tree estimation.