Phylogenetic signal of genomic repeat abundances can be distorted by random homoplasy: a case study from hominid primates

The genomic abundance of different types of repetitive DNA elements contains a phylogenetic signal useful for inferring the evolutionary history of different groups of organisms. Here we test the reliability of this approach using the Hominidae family of primates, whose consensus phylogeny is well accepted. We used the software RepeatExplorer to identify the different repetitive DNA clusters and quantify their abundances. With these data, we performed phylogenetic analyses by maximum parsimony, including one, two or three individuals per species, technical replicates, and including or discarding two clusters of repetitive elements (i.e. A satellite DNA and an endogenous retrovirus) that generated random homoplasy, because they were abundant in Pan and Gorilla but almost absent in Homo and Pongo. The only phylogenetic tree congruent with the accepted topology for hominids, thus coinciding with that obtained from the mitogenomes of the same individuals, was the one built after filtering out the libraries for the two homoplasious clusters and using three individuals per species. Our results suggest some caution in the use of repeat abundance for phylogenetic studies, because some element abundances are homoplasious, which severely distorts the phylogenetic signal owing to their differential amplification among evolutionary lineages.