How stable is the "Polyphyly of Lice" hypothesis (Insecta: Psocodea)?: a comparison of phylogenetic signal in multiple genes.

Recent molecular phylogenetic analyses of 18S rDNA have indicated that parasitic lice (order Phthiraptera) are possibly polyphyletic. These analyses recovered one of the parasitic louse suborders, Amblycera, as the sister group to the free-living booklouse family Liposcelididae. We further tested this hypothesis using DNA sequences from five genes: nuclear 18S rDNA, Histone 3, and wingless and mitochondrial 16S rDNA and COI. Combined analyses of these five genes provided reasonably strong support for the Amblycera+Liposcelididae clade, supporting the polyphyly of lice hypothesis. To explore the robustness of this result, we examined the phylogenetic signal contained in each gene independently (except for wingless, which could not be readily amplified in many target taxa). Analyses of each gene separately and in various combinations with other genes revealed that clear signal supporting Amblycera+Liposcelididae only existed in the 18S data, although no analysis supported monophyly of parasitic lice. Nevertheless, combined analyses of all genes provided stronger support for this relationship than that obtained from 18S data alone. The increase in support for this clade was mostly explained by the stabilization of other parts of the tree and potentially inappropriate substitution modeling. These findings demonstrate that the increased support values provided by combined data set does not always indicate corroboration of the hypothesis. 2010 Elsevier Inc. All rights reserved.