Rerooting the rDNA gene tree reveals phoronids to be ‘brachiopods without shells’; dangers of wide taxon samples in metazoan phylogenetics (Phoronida; Brachiopoda)

Abstract

Molecular phylogenetics has resulted in conflicting accounts of the relationship between phoronids and brachiopods. Taxonomically comprehensive analyses of brachiopod and phoronid ribosomal DNA sequences (rDNAs) rooted with short-branched mollusc sequences uniformly find that phoronids nest within brachiopods as the sister of the three extant inarticulate lineages. Here, this is called the ‘alternate’ topology because it does not match traditional, morphology-based ideas. Many other analyses of protein-coding genes and/or rDNAs place phoronids elsewhere, often as the sister group of all brachiopods, better matching ‘traditional’ ideas. However, these analyses generally are based on data from small selections of brachiopods and phoronids, include data from a wide range of other metazoan taxa, and are rooted with distant outgroups. Here, I show that outgroup rooting of brachiopods and phoronid rDNAs is unreliable, and instead find the root position with procedures that are free from all distortions caused by distantly related taxa, i.e. by Bayesian and maximum likelihood relaxed-clock analyses of a purely ingroup alignment. All such analyses confirm the ‘alternate’ topology: phoronids belong within the Brachiopoda as the sister group of the inarticulates. In addition, nine factors are identified that (singly or in combination) can cause misreporting of the phylogenetic signal in wide taxon-range analyses of both rDNA and amino acid sequence data.