Skip to main content
Dryad

Effect of different types of sequence data on palaeognath phylogeny

Takezaki, Naoko1

Published Mar 28, 2023; Updated Apr 04, 2023 on Dryad. https://doi.org/10.5061/dryad.1jwstqjzs

Data files

Mar 28, 2023 version files 731.03 MB

Abstract

Palaeognathae consists of five groups of extant species: flighted tinamous (1) and four flightless groups: kiwi (2), cassowaries and emu (3), rheas (4), and ostriches (5). Molecular studies supported the groupings of extinct moas with tinamous and ele- phant birds with kiwi as well as ostriches as the group that diverged first among the five groups. However, phylogenetic re- lationships among the five groups are still controversial. Previous studies showed extensive heterogeneity in estimated gene tree topologies from conserved nonexonic elements, introns, and ultraconserved elements. Using the noncoding loci to- gether with protein-coding loci, this study investigated the factors that affected gene tree estimation error and the relation- ships among the five groups. Using closely related ostrich rather than distantly related chicken as the outgroup, concatenated and gene tree–based approaches supported rheas as the group that diverged first among groups (1)–(4). Whereas gene tree estimation error increased using loci with low sequence divergence and short length, topological bias in estimated trees oc- curred using loci with high sequence divergence and/or nucleotide composition bias and heterogeneity, which more occurred in trees estimated from coding loci than noncoding loci. Regarding the relationships of (1)–(4), the site patterns by parsimony criterion appeared less susceptible to the bias than tree construction assuming stationary time-homogeneous model and sug- gested the clustering of kiwi and cassowaries and emu the most likely with ∼40% support rather than the clustering of kiwi and rheas and that of kiwi and tinamous with 30% support each.