[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]
Turtle phylogenetic position based on genes as characters
From: Ben Creisler
bcreisler@gmail.com
New in PLoS ONE:
Bin Lu, Weizhao Yang, Qiang Dai & Jinzhong Fu (2013)
Using Genes as Characters and a Parsimony Analysis to Explore the
Phylogenetic Position of Turtles.
PLoS ONE 8(11): e79348.
doi:10.1371/journal.pone.0079348
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0079348
The phylogenetic position of turtles within the vertebrate tree of
life remains controversial. Conflicting conclusions from different
studies are likely a consequence of systematic error in the tree
construction process, rather than random error from small amounts of
data. Using genomic data, we evaluate the phylogenetic position of
turtles with both conventional concatenated data analysis and a “genes
as characters” approach. Two datasets were constructed, one with seven
species (human, opossum, zebra finch, chicken, green anole, Chinese
pond turtle, and western clawed frog) and 4584 orthologous genes, and
the second with four additional species (soft-shelled turtle, Nile
crocodile, royal python, and tuatara) but only 1638 genes. Our
concatenated data analysis strongly supported turtle as the
sister-group to archosaurs (the archosaur hypothesis), similar to
several recent genomic data based studies using similar methods. When
using genes as characters and gene trees as character-state trees with
equal weighting for each gene, however, our parsimony analysis
suggested that turtles are possibly sister-group to diapsids,
archosaurs, or lepidosaurs. None of these resolutions were strongly
supported by bootstraps. Furthermore, our incongruence analysis
clearly demonstrated that there is a large amount of inconsistency
among genes and most of the conflict relates to the placement of
turtles. We conclude that the uncertain placement of turtles is a
reflection of the true state of nature. Concatenated data analysis of
large and heterogeneous datasets likely suffers from systematic error
and over-estimates of confidence as a consequence of a large number of
characters. Using genes as characters offers an alternative for
phylogenomic analysis. It has potential to reduce systematic error,
such as data heterogeneity and long-branch attraction, and it can also
avoid problems associated with computation time and model selection.
Finally, treating genes as characters provides a convenient method for
examining gene and genome evolution.