New modern bird phylogenies

[David Černý <david.cerny1@gmail.com>]

Two new neontological papers that might be of interest but have not
been mentioned yet on the list:

Kimball RT, Wang N, Heimer-McGinn V, Ferguson C, Braun EL 2013
Identifying localized biases in large datasets: A case study using the
Avian Tree of Life. Mol Phylogenet Evol
doi:10.1016/j.ympev.2013.05.029
http://www.sciencedirect.com/science/article/pii/S1055790313002431

Large-scale multi-locus studies have become common in molecular
phylogenetics, with new studies continually adding to previous
datasets in an effort to fully resolve the tree of life. Total
evidence analyses that combine existing data with newly collected data
are expected to increase the power of phylogenetic analyses to resolve
difficult relationships. However, they might be subject to localized
biases, with one or a few loci having a strong and potentially
misleading influence upon the results. To examine this possibility we
combined a newly collected 31-locus dataset that includes
representatives of all major avian lineages with a published dataset
of 19 loci that has a comparable number of sites (Hackett et al. 2008
Science 320, 1763-1768). This allowed us to explore the advantages of
conducting total evidence analyses, and to determine whether it was
also important to analyze new datasets independent of published ones.
The total evidence analysis yielded results very similar to the
published results, with only slightly increased support at a few
nodes. However, analyzing the 31- and 19-locus datasets separately
highlighted several differences. Two clades received strong support in
the published dataset and total evidence analysis, but the support
appeared to reflect bias at a single locus (β-fibrinogen [FGB]). The
signal in FGB that supported these relationships was sufficient to
result in their recovery with bootstrap support, even when combined
with 49 loci lacking that signal. FGB did not appear to have a
substantial impact upon the results of species tree methods, but
another locus (brain-derived neurotrophic factor [BDNF]) did have an
impact upon those analyses. These results demonstrated that localized
biases can influence large-scale phylogenetic analyses but they also
indicated that considering independent evidence and exploring multiple
analytical approaches could reveal them.


The overall methodology is similar to several previous papers by the
same group of researchers (Hackett et al. 2008; Wang et al. 2011;
Smith et al. 2012) and includes species-tree analyses,
gene-jackknifing to identify "rogue" genes, and measures against
alignment bias. The 31 new loci produce a tree that is quite congruent
with Hackett et al.'s results, although the topology of landbirds is
different -- hawks, mousebirds, and the cuckoo roller are more closely
related to passerines than to woodpeckers. Shorebirds are closer to
the "waterbird assemblage" (Aequornithes) than to landbirds (unlike in
Hackett al. 2008 and retroposon studies) and eurypygiforms (the
kagu/sunbittern grouping) are gruiforms, as in old morphological
analyses and traditional classifications. The
Eurypygiformes/Gruiformes clade forms a sister group to tropicbirds,
which is similar to what McCormack et al. (2013) recovered using
ultraconserved elements. Interestingly, there is a clade resembling
Metaves: it includes some former "coronavians" (turacos, bustards) and
excludes a few supposedly metavian taxa (pigeons, eurypygiforms,
tropicbirds). The 50-locus supermatrix supports a topology that is
even more in agreement with Hackett et al. (2008), although there is
one big, rather unlikely difference -- loons aren't sister to all
other aequornithines but to a clade comprising pelicans and
(nonmonophyletic) Suliformes. Without the notorious FGB-int7, all
"metavians" except eurypygiforms and tropicbirds (which are together
sister to a clade comprising core-gruiforms and landbirds) form a
series of outgroups to all other neoavians, with a sandgrouse/mesite
clade being most basal and the hoatzin most derived.


Yuri T, Kimball RT, Harshman J, Bowie RCK, Braun MJ, Chojnowski JL,
Han K-L, Hackett SJ, Huddleston CJ, Moore WS, Reddy S, Sheldon FH,
Steadman DW, Witt CC, Braun EL 2013 Parsimony and model-based analyses
of indels in avian nuclear genes reveal congruent and incongruent
phylogenetic signals. Biology 2: 419–44
http://www.mdpi.com/2079-7737/2/1/419 (open access!)

Insertion/deletion (indel) mutations, which are represented by gaps in
multiple sequence alignments, have been used to examine phylogenetic
hypotheses for some time. However, most analyses combine gap data with
the nucleotide sequences in which they are embedded, probably because
most phylogenetic datasets include few gap characters. Here, we report
analyses of 12,030 gap characters from an alignment of avian nuclear
genes using maximum parsimony (MP) and a simple maximum likelihood
(ML) framework. Both trees were similar, and they exhibited almost all
of the strongly supported relationships in the nucleotide tree,
although neither gap tree supported many relationships that have
proven difficult to recover in previous studies. Moreover, independent
lines of evidence typically corroborated the nucleotide topology
instead of the gap topology when they disagreed, although the number
of conflicting nodes with high bootstrap support was limited.
Filtering to remove short indels did not substantially reduce
homoplasy or reduce conflict. Combined analyses of nucleotides and
gaps resulted in the nucleotide topology, but with increased support,
suggesting that gap data may prove most useful when analyzed in
combination with nucleotide substitutions.


ML analyses support the Metaves hypothesis, but the rest of the tree
looks weird; landbirds are nested in Aequornithes and their topology
is all wrong (woodpeckers, rollers, and hornbills are recovered as
eufalconimorphs). Another interesting thing is that the authors
present a comprehensive novel nomenclature for higher-level neoavian
clades, including many groups that have been robustly supported
previously but not named (e.g., the "landbird assemblage"). Landbirds
are now called "Telluraves", non-ostrich paleognaths
"Notopalaeognathae". Recent names such as Psittacopasserae are
retained, even though the supplementary info contains a half-page long
rant about Eufalconimorphae and Ericson's name for the
seriema/falcon/parrot/passerine clade is corrected from "Australavis"
to Australaves. Fans of phylogenetic nomenclature might be pleased
with the fact that the authors define all these names, though in a way
which is a bit lazy and probably not PhyloCode-compliant (as "the
least inclusive clade comprising the relevant species in the Early
Bird tree").


*Refs:*

Hackett SJ, Kimball RT, Reddy S, Bowie RC, Braun EL, Braun MJ,
Chojnowski JL, Cox WA, Han K, Harshman J, Huddleston CJ, Marks BD,
Miglia KJ, Moore WS, Sheldon FH, Steadman DW, Witt CC, Yuri T 2008 A
phylogenomic study of birds reveals their evolutionary history.
Science 320(5884): 1763–8

McCormack JE, Harvey MG, Faircloth BC, Crawford NG, Glenn TC,
Brumfield RT 2013 A phylogeny of birds based on over 1,500 loci
collected by target enrichment and high-throughput sequencing. PLoS
ONE 8(1): e54848

Smith JV, Braun EL, Kimball RT 2012 Ratite non-monophyly: Independent
evidence from 40 novel loci. Syst Biol 62(1): 35–49

Wang N, Braun EL, Kimball RT 2011 Testing hypotheses about the sister
group of the Passeriformes using an independent 30-locus data set. Mol
Biol Evol 29(2): 737–50


-- 
David Černý