Zoologica Scripta
Invited Review
The phylogenomic forest of bird trees contains a hard
polytomy at the root of Neoaves
ALEXANDER SUH
Submitted: 11 April 2016
Accepted: 11 August 2016
doi:10.1111/zsc.12213
Suh, A. (2016). The phylogenomic forest of bird trees contains a hard polytomy at the root
of Neoaves. — Zoologica Scripta, 45, 50–62.
Birds have arguably been the most intensely studied animal group for their phylogenetic
relationships. However, the recent advent of genome-scale phylogenomics has made the forest of bird phylogenies even more complex and confusing. Here, in this perspective piece, I
show that most parts of the avian Tree of Life are now firmly established as reproducible
phylogenetic hypotheses. This is to the exception of the deepest relationships among
Neoaves. Using phylogenetic networks and simulations, I argue that the very onset of the
super-rapid neoavian radiation is irresolvable because of eight near-simultaneous speciation
events. Such a hard polytomy of nine taxa translates into 2 027 025 possible rooted bifurcating trees. Accordingly, recent genome-scale phylogenies show extremely complex conflicts
in this (and only this) part of the avian Tree of Life. I predict that the upcoming years of
avian phylogenomics will witness many more, highly conflicting tree topologies regarding
the early neoavian polytomy. I further caution against bootstrapping in the era of genomics
and suggest to instead use reproducibility (e.g. independent methods or data types) as support for phylogenetic hypotheses. The early neoavian polytomy coincides with the Cretaceous–Paleogene (K-Pg) mass extinction and is, to my knowledge, the first empirical
example of a hard polytomy.
Corresponding author: Alexander Suh, Department of Evolutionary Biology, Evolutionary Biology
Centre (EBC), Uppsala University, SE - 752 36, Uppsala, Sweden. E-mail: alexander.suh@
ebc.uu.se
Alexander Suh, Department of Evolutionary Biology, Evolutionary Biology Centre (EBC), Uppsala
University, SE - 752 36, Uppsala, Sweden. E-mail: [email protected]
Introduction
The last few years have produced so many different phylogenetic trees of birds that it has become almost impossible
to see the forest for the trees. It is thus about time to revisit
recent contradictory claims from genome-scale studies and
to disentangle the degrees of conflict present in different
parts of the avian Tree of Life. Nearly all of these conflicts
involve the super-rapid radiation of Neoaves, the taxon
comprising all extant birds but the lineages leading to chickens, ducks, ostriches, cassowaries and tinamous. The
Neoaves problem has been intensely debated for decades. In
the following, I show that genome-scale phylogenomics has
resolved most of these disputes, and I argue that the
presence of a hard polytomy explains all of the remaining
irresolvable branches of the neoavian Tree of Life.
To test for a hard polytomy, I generated phylogenetic
networks in Splitstree4 (Huson & Bryant 2006) using
50
publicly available trees/data sets (Jarvis et al. 2014, 2015;
Prum et al. 2015; Suh et al. 2015). This was complemented
with networks based on simulated nine-taxon gene trees
from a random tree generator (Boc et al. 2012) and simulated nine-taxon retroposon presence/absence patterns (i.e.
all 512 possible strings of nine binary digits).
The emerging consensus of a largely resolved
neoavian Tree of Life
There have been numerous attempts to solve the Neoaves
problem. Each of these has seen extensive disagreement
between and among studies, irrespective of whether the
data type was morphology (e.g. Livezey & Zusi 2007),
DNA–DNA hybridization (e.g. Sibley & Ahlquist 1990),
mitochondrial genes (e.g. Pratt et al. 2009; Pacheco et al.
2011) or single nuclear genes (e.g. Fain & Houde 2004;
Poe & Chubb 2004). If one were to draw a strict consensus
ª 2016 The Authors. Zoologica Scripta published by John Wiley & Sons Ltd on behalf of Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
This is an open access article under the terms of the Creative Commons Attribution License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
A. Suh
of all these analyses, it would be an entirely unresolved
neoavian ‘Bush of Life’.
Over the last ten years, this picture has changed dramatically with analyses of more and more nuclear loci. While
exons have been shown to be problematic for resolving the
neoavian radiation (Chojnowski et al. 2008; Jarvis et al.
2014), other nuclear data types such as introns, ultraconserved elements (UCEs) and retroposon presence/absence
patterns have become increasingly important. Analysing
even just a few handfuls of unlinked nuclear loci has
revealed some surprising relationships that since then have
been recovered relatively consistently, such as Psittacopasserae (passerines + parrots), Mirandornithes (flamingos
+ grebes), Telluraves (core landbirds) and Aequornithes
(core waterbirds) (van Tuinen et al. 2001; Ericson et al.
2006; Hackett et al. 2008; Suh et al. 2011, 2012; Wang
et al. 2012). Other, equally surprising relationships have
only been discovered through analysis of hundreds or thousands of unlinked nuclear loci, such as Phaethontimorphae
(sunbittern + tropicbirds), the turaco/bustard clade and the
sandgrouse/mesite clade (McCormack et al. 2013; Jarvis
et al. 2014; Prum et al. 2015; Suh et al. 2015).
Since the end of 2015, it has been safe to say that there
is now a new and robust consensus for the neoavian Tree
of Life (Fig. 1). With the exception of the eight deepest
speciation events of the neoavian radiation, there appear to
be reproducible phylogenetic hypotheses for all other longstanding aspects of the Neoaves problem. I emphasize that
this reproducibility lies not in high values for bootstrap or
Bayesian posterior probabilities, the reasons for which will
be discussed below. Instead, the new neoavian Tree of Life
is largely robust and reproducible because most relationships have been recovered in at least two phylogenetic trees
based on independent methods or data types (Fig. 1).
These are four genome-scale phylogenies, namely the Jarvis et al. (2014) TENT tree (48 taxa, 14 536 nuclear loci),
the McCormack et al. (2013) tree (32 taxa, 1541 UCE loci),
the Prum et al. (2015) tree (198 taxa, 259 nuclear loci) and
the Suh et al. (2015) tree (48 taxa, 2118 retroposon loci).
For comparison in Fig. 1, I also included the Jarvis et al.
(2014) UCE tree (48 taxa, 3769 UCE loci) and the historically important and widely known Hackett et al. (2008) tree
(169 taxa, 19 intron loci).
What is the lesson to be learnt from all this? It is not
simply the addition of more data, but the replication of
such analyses via independent data types, independent
taxon samplings and different teams of researchers, that
lends credibility to the new consensus of the avian Tree of
Life. Not only does this apply to the aforementioned relationships that have now been recovered consistently in
phylogenomic analyses, but also to some clades that were
newly found in at least two of the four genome-scale
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A hard polytomy at the root of Neoaves
studies (McCormack et al. 2013; Jarvis et al. 2014; Prum
et al. 2015; Suh et al. 2015), for example Otidimorphae
(bustards + turacos + cuckoos), Columbimorphae (mesites
+ sandgrouse + doves), the Aequornithes/Phaethontimorphae clade and Afroaves (core landbirds excluding Australaves) (Fig. 1).
Some neoavian relationships are more complex than
a bifurcating tree
It is relieving that there are now robust and reproducible
phylogenetic hypotheses for most aspects of the Neoaves
problem (Fig. 1). But genome-scale phylogenomics
notwithstanding, why do there continue to be controversial
branches at the base of core landbirds and unresolved
branches at the onset of the neoavian radiation? As so often
in biology, things are a bit more complex than they seem.
First of all, the current mainstream approach of measuring and comparing ‘support’ in genome-scale phylogenomics may be flawed, as nicely illustrated by the recent
example of the enigmatic hoatzin. In their Nature paper,
Prum et al. (2015) stated that their ‘[. . .] concatenated Bayesian analyses resulted in a completely resolved, well supported
phylogeny. [. . .] Almost all clades in the maximum likelihood
tree were maximally supported with bootstrap scores (BS) of
1.00’. They used this ‘support’ to group the hoatzin with
core landbirds and name the ‘novel’ clade ‘Inopinaves’
(Fig. S1B). However, the same argument was used in the
Science paper by Jarvis et al. (2014) to group the hoatzin
with cranes + shorebirds and name the ‘novel’ clade
‘Gruae’ (Fig. S1A). How can we trust bootstrap or Bayesian posterior probabilities if different trees exhibit discordant topologies that all have maximum ‘support’? Salichos
& Rokas (2013) foresaw this problem a few years ago
when they stated that ‘[. . .] relying on bootstrap to analyse
phylogenomic data sets is misleading, [. . .] because its application
will, even in the presence of notable conflict or systematic error,
almost always result in 100% values’. Genome-scale phylogenomics thus urgently needs a paradigm shift, away from
overselling ‘novel’ clades and instead defining support as
reproducibility of a phylogenetic hypothesis – that is, the
emergence of the same bifurcation in independent data
sets analysed by independent research groups using independent methods. Additionally, I hope that measures such
as internode certainty (Salichos & Rokas 2013), now
implemented in RAxML (Salichos et al. 2014), will soon
replace bootstrap and conventional Bayesian posterior
probabilities in genome-scale phylogenomics. A recent
study on complex speciation in tomatoes shows that this is
feasible (Pease et al. 2016). Furthermore, Bayesian analyses
using a reversible-jump Markov chain Monte Carlo
approach yield more reliable Bayesian posterior probabilities than conventional Bayesian analyses under uniform
51
A hard polytomy at the root of Neoaves
A. Suh
Fig. 1 The new consensus of the avian Tree of Life. Topological support by six different trees is colour-coded. Branches recovered by less
than two independent studies are collapsed and white boxes denote conflicting topologies at the respective node. Higher-level names follow
the nomenclature in Jarvis et al. (2014), with the exception of those taxon names (asterisks) that were described in prior studies (Sangster
2005; Mayr 2011; Suh et al. 2011; Yuri et al. 2013).
52
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A. Suh
branch length priors (Lewis et al. 2005; Yang & Rannala
2005; Yang 2007).
Genome-scale phylogenomics requires a second paradigm shift. For reasons mentioned above, the addition of
more and more data yields increasingly ‘resolved’ species
trees (Salichos & Rokas 2013). At the same time, sampling
more genomic loci increases the variance among the gene
trees that underlie the species tree, leading to more, not
less, phylogenetic discordance (Jeffroy et al. 2006; Philippe
et al. 2011; Liu et al. 2015; Hahn & Nakhleh 2016). The
most striking example for this is the hitherto largest genome-scale analysis of any animal group, the Jarvis et al.
(2014) TENT tree in which not a single one of the 14 536
individual gene trees is identical to the species tree. Furthermore, gene trees from different data types differ from
the species tree on most branches (Jarvis et al. 2014). This
dilemma illustrates that ‘resolved’ species trees are a problematic metaphor for genome-scale data because they hide
ubiquitous discordance among gene trees (Hahn & Nakhleh 2016). Thus, I hope that in the near future every analysis of bifurcating trees will be complemented by
phylogenetic networks, because these visualize alternative
topologies as reticulations (Whitfield & Lockhart 2007;
Hallstr€
om & Janke 2010; Bapteste et al. 2013). One such
analysis of retroposon presence/absence patterns revealed
that Neoaves in fact comprise three rapid radiations,
namely an initial super-radiation, the core waterbird radiation and the core landbird radiation (Fig. 4A of Suh et al.
2015). The neighbour network analysis also showed that all
the controversial and unresolved parts of the neoavian Tree
of Life (cf. Fig. 1) contain high degrees of phylogenetic
discordance (Suh et al. 2015). As carefully analysed retroposon presence/absence patterns contain negligible
amounts of homoplasy (Shedlock et al. 2004; Ray et al.
2006; Han et al. 2011), these discordances have been attributed to an extreme prevalence of incomplete lineage sorting (ILS) in some parts of the neoavian Tree of Life (Suh
et al. 2015). ILS, the persistence of ancestral polymorphisms across multiple speciation events, is ubiquitous in
nature (Maddison & Knowles 2006). Subsequent fixation of
one of the two alleles of a polymorphism occurs independently in each descendant lineage, leading to discordances
that have been dubbed ‘hemiplasy’ because, unlike homoplasy, the alleles are identical by descent (Avise & Robinson 2008). ILS is particularly prevalent when effective
population size (Ne) is large and time between consecutive
speciation events is short (Maddison 1997; Maddison &
Knowles 2006). It is thus interesting that there are very
high levels of ILS at the respective onsets of the initial
neoavian super-radiation and the core landbird radiation,
followed by a gradual slowdown of ILS towards the tips of
the neoavian Tree of Life (Suh et al. 2015).
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A hard polytomy at the root of Neoaves
The phylogenetic network approach can be extended to
compare gene trees or species trees from different studies.
A consensus network (Fig. 2) based on three species trees
from the largest genome-scale analyses to date (Jarvis et al.
2014; Prum et al. 2015; Suh et al. 2015) largely resembles
the neoavian Tree of Life (cf. Fig. 1, cf. Fig. S2A). Reticulations are restricted to the controversial branches at the
onset of the core landbird radiation and the unresolved
branches at the onset of the neoavian radiation. Below, I
address these two issues separately because they constitute
two different phylogenetic problems.
Why is the root of core landbirds controversial?
The controversial branches at the onset of the core landbird radiation result from changes in the respective positions of Accipitriformes (eagles + NW vultures),
Coliiformes (mousebirds) and Strigiformes (owls) (Fig. 2,
Fig. 3A–B). However, the majority of independent genome-scale species trees have grouped Accipitriformes and
owls as illustrated in Fig. 1 (McCormack et al. 2013; Jarvis
et al. 2014; Suh et al. 2015). The only remaining controversy is the mousebird problem with two recurrent alternative positions (Suh et al. 2015). In analyses that include
exons and/or introns, mousebirds have grouped within
Coraciimorphae (Jarvis et al. 2014; Prum et al. 2015) as sister to Cavitaves (woodpeckers + bee-eaters + hornbills +
cuckoo roller; Yuri et al. 2013). However, independent data
sets of UCEs or retroposon markers have consistently
recovered mousebirds as sister to the remaining Afroaves
(Jarvis et al. 2014; Suh et al. 2015). A possible explanation
for this discordance would be strong genomic signal for
the two competing hypotheses, resulting from introgressive
hybridization (i.e. postspeciation gene flow) between the
ancestor of mousebirds and the ancestor of Cavitaves.
However, tests of retroposon support suggested that
hybridization did not play a measurable role in the mousebird problem (Suh et al. 2015), and the only prevalent
genomic signal is that for the grouping of mousebirds as
sister to the remaining Afroaves (Fig. 3C). I suggest an
alternative explanation for the grouping of mousebirds with
Cavitaves in genome-scale analyses involving exons and
introns. Compared to the more slowly evolving raptorial
lineages within Afroaves, mousebirds and Cavitaves have
longer branches (cf. Fig. 3 of Hackett et al. 2008 and
Fig. S7 of Jarvis et al. 2014), raising suspicion of longbranch attraction (LBA) in sequence analyses of exons and
introns. This would explain the previous notion that
mousebirds are a ‘rogue’ taxon in such analyses (Wang
et al. 2012). It remains to be tested whether minimizing
LBA in existing intron/exon data (e.g. via the LS3 method
of Rivera-Rivera & Montoya-Burgos 2016) might overcome the rogue effect of mousebirds and recover them as
53
A hard polytomy at the root of Neoaves
A. Suh
Fig. 2 The avian Tree of Life contains local reticulations and a polytomy zone. Average consensus network based on the three largest,
independent genome-scale phylogenies to date (Jarvis et al. 2014; Prum et al. 2015; Suh et al. 2015). The Prum et al. (2015) tree was
pruned to match the taxon sampling of the Jarvis et al. (2014) and Suh et al. (2015) trees (see Fig. 1 for information about methods and
data types employed in the three studies and Figure S2A for a majority-rule consensus tree). The polytomy zone is marked in yellow and
the remaining colour coding corresponds to the three neoavian radiations defined by Suh et al. (2015), namely the initial superradiation
(red), the core waterbird radiation (blue) and the core landbird radiation (green).
sister to the remaining Afroaves, consistent with UCEs and
retroposon markers.
Why is the root of Neoaves unresolved?
The unresolved relationships at the very onset of the neoavian radiation (Fig. 1) perfectly overlap with those branches
that Suh et al. (2015) recently found to exhibit ‘[. . .] extreme
54
level[s] of up to 100% ILS [in retroposon markers] [. . .] [creating arrangements that] may neither resemble a fully bifurcating
tree nor are they resolvable as such’. Although this notion was
ignored in the subsequent phylogenomic study of Prum
et al. (2015), their ‘resolved’ tree in fact further strengthens
the aforementioned suggestion that the root of Neoaves is
not resolvable. Not only are the major tree hypotheses of
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A. Suh
A hard polytomy at the root of Neoaves
Fig. 3 The onset of the core landbird radiation (i.e. the mousebird problem) is not a hard polytomy. All avian phylogenetic networks
were pruned to one representative for each of the five taxa (pictured). (A) Pruned version of fig. 5D of Suh et al. (2015) (B) Pruned
version of Fig. 2 of this study. (C) Pruned version of Fig. 4A of Suh et al. (2015). Consensus networks (panels A–B) show consensus
splits present in at least 5% of trees and the neighbour network (panel C) shows reticulations among 0/1-coded character
distributions. The five taxa involved in the mousebird problem are reciprocally monophyletic (cf. Figs 1–2), namely Accipitriformes
(Ac), Australaves (Au), Cavitaves (Cv), Coliiformes (Ci) and Strigiformes (St). Colour coding of phylogenetic networks corresponds to
Fig. 2.
Jarvis et al. (2014) and Prum et al. (2015) entirely discordant with regard to the eight deepest neoavian relationships
(Fig. S1), but these conflicts are additionally so complex
that they collapse into an unresolved ‘tangle’ in attempts to
reconcile them (Fig. 2, Fig. 4C, Fig. 2A). It is important to
keep in mind that this star-like tangle is different from the
previously discussed mousebird problem, which exhibits
tree-like reticulations that are much less complex and
resolvable as a bifurcating species tree (cf. Fig. 4A–D vs.
Fig. 3).
What is this early neoavian tangle? Previous studies in
the pregenomic era have repeatedly argued for (Poe &
Chubb 2004) or against (Gibb et al. 2007; Chojnowski et al.
2008) a polytomy within Neoaves. In this context, it is
important to distinguish between two types of polytomies.
In contrast to ‘soft polytomies’ that result from insufficient
data, ‘hard polytomies’ reflect the biological limit of phylogenetic resolution because of near-simultaneous speciation
(Hoelzer & Meinick 1994; Walsh et al. 1999; Whitfield &
Lockhart 2007). Now that genome assemblies are available
for all major lineages of Neoaves (Jarvis et al. 2014; Zhang
et al. 2014), there are finally sufficient data to rule out the
presence of soft polytomies via genome-scale phylogenomics. To do so, I conducted phylogenetic network analyses of existing data sets derived from 48 bird genomes. The
results can be considered minimum estimates for phylogenetic conflict, because an increase in taxon sampling is
likely to recover more discordances due to ILS. I analysed
3679 UCE gene trees and 2022 binned intron gene trees
(Jarvis et al. 2014, 2015) for the presence of reticulations
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
present in at least 5% of the respective trees. The resultant
consensus networks generally exhibit tree-like reticulations,
with the exception of the root of Neoaves (Fig. S3, cf.
Fig. 2). Illustrating the extent of the conflict between genome-scale species trees (Figs 1–2), a total of nine reciprocally monophyletic taxa originate in this unresolved tangle
at the root of Neoaves. Pruning networks to one representative for each of these nine taxa again reveals a star-like
tangle (Fig. 4A–B). The same star-like signal was previously noted in a neighbour network analysis of low-homoplasy retroposon markers (Fig. 4A of Suh et al. 2015) and
persists when pruning this network to one representative
for each of the nine critical taxa (Fig. 4D). Thus, neither
adding more data nor reducing homoplasy decreases the
complexity at the root of Neoaves. On the contrary, irrespective of whether one looks at genome-scale species
trees, individual gene trees or low-homoplasy retroposon
presence/absence patterns (Fig. 4), the tangle persists. This
suggests a lack of underlying tree-like signal. I thus argue
that the very onset of the neoavian radiation was characterized by eight near-simultaneous speciation events that led
to a nine-taxon hard polytomy.
How can we test the hypothesis of a nine-taxon hard
polytomy at the root of Neoaves? Slowinski (2001)
suggested that ‘[. . .] when an internode on a species tree is
zero-length, each possible gene tree corresponding to each possible
resolution of the species polytomy is equiprobable’. It is possible
to determine the number of possible gene trees for nine
taxa following the formula by Cavalli-Sforza & Edwards
(1967). Accordingly, 2 027 025 rooted bifurcating trees are
55
A hard polytomy at the root of Neoaves
A. Suh
Fig. 4 The onset of the neoavian radiation is a nine-taxon hard polytomy. All avian phylogenetic networks were pruned to one
representative for each of the nine taxa (pictured). (A) Pruned network of 3,679 UCE gene trees from Jarvis et al. (2014) (cf. Fig. S3A), (B)
pruned network of 2022 binned intron gene trees from Jarvis et al. (2014) (cf. Fig. S3B), (C) pruned version of Fig. 2 of this study. (D)
Pruned version of Fig. 4A of Suh et al. (2015), (E) simulated nine-taxon polytomy based on a sample of 1000 random gene trees and (F)
simulated nine-taxon polytomy based on a random sample of 95% of the 512 possible retroposon presence/absence patterns (cf. Fig. S2B).
Consensus networks (panels A–C, panel E) show consensus splits present in at least 5% of trees whereas neighbour networks (panels D and
G) show reticulations among 0/1-coded character distributions. The nine taxa originating in this polytomy zone are reciprocally
monophyletic (cf. Figs 1–2, Fig. S1), namely Aequornithes/Phaethontimorphae (Ae/Ph), Caprimulgiformes (Ca), Charadriiformes (Ch),
Columbimorphae (Co), Gruiformes (Gr), Opisthocomiformes (Op), Mirandornithes (Mi), Otidimorphae (Ot) and Telluraves (Te). Colour
coding of phylogenetic networks corresponds to Fig. 2.
possible (Felsenstein 1978) and these should be equiprobable under the null hypothesis of a nine-taxon hard polytomy. I therefore simulated a nine-taxon hard polytomy by
generating a random set of 1000 gene trees for nine taxa.
The resultant consensus network (Fig. 4E) is strikingly
similar to the aforementioned star-like tangle among
empirical gene trees (Fig. 4A–B). I propose an additional
test of this null hypothesis by simulating the data underlying individual phylogenetic trees. Under the scenario of a
hard polytomy giving rise to n taxa, all of the shared ancestral polymorphisms will undergo ILS and subsequent
stochastic sorting in each of the n descendant lineages.
56
Assuming that these ancestral polymorphisms are biallelic
markers (e.g. single-nucleotide polymorphisms or retroposon presence/absence polymorphisms), this translates
into 2n possible character distributions which should be
equiprobable under the null hypothesis of a hard polytomy.
Consequently, a nine-taxon polytomy leads to 512 possible
character distributions. Following the formula of Suh et al.
(2015), 96.5% of these show hemiplasy, which means that
this percentage of character distributions is discordant with
each of the 2 027 025 possible bifurcating species tree
topologies. I thus simulated retroposon presence/absence
patterns by picking a random subset of 95% of all 512
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A. Suh
possible strings of nine binary digits. Neighbour network
analysis again reveals star-like signal (Fig. 4F, cf. Fig. S2B)
that is strikingly similar to the empirical retroposon data
(Fig. 4D). Taken together, the analysed empirical and simulated data suggest that the root of Neoaves is a nine-taxon
hard polytomy.
Is it possible to invoke hybridization as a full or partial
explanation of this hard polytomy-like pattern? According
to the ‘hybrid swarm theory’, hybridization may spur
rapid adaptive diversification (Seehausen 2004). In this
context, it is important to distinguish between gene flow
during speciation (natural hybridization) and postspeciation gene flow (introgressive hybridization) (Harrison &
Larson 2014). Gene flow during speciation is virtually
indistinguishable from ILS and adds to the overall extent
and duration of ILS (Suh et al. 2015). On the other
hand, postspeciation gene flow can be distinguished from
ILS, and there are now a number of methods to do so
for single-nucleotide polymorphisms (e.g. Durand et al.
2011; Martin et al. 2015; Pease & Hahn 2015) and retroposon presence/absence patterns (Kuritzin et al. 2016).
However, these phylogenetic tree methods handle trees
with only four or five taxa and require a known species
tree. New methods are therefore necessary for analysing
complex phylogenetic scenarios which involve more than
five taxa and lack a ‘resolved’ species tree. Alternatively,
it remains to be tested whether existing phylogenetic network methods (e.g. Yu et al. 2013, 2014; Solıs-Lemus &
Ane 2016) can handle the aforementioned complexity of
the Neoaves problem. At any rate, if the star-like tangle
at the root of Neoaves was caused by introgressive
hybridization rather than a hard polytomy, one would
have to assume near-equal frequencies of postspeciation
hybridization among nine species. This scenario seems
much less plausible than assuming a hard polytomy due
to nine near-simultaneous speciation events at the onset
of the neoavian radiation.
The explosive super-radiation at the onset of neoavian
diversification can thus be described as a polytomy zone.
How does the concept of a polytomy zone relate to
another concept, the ‘anomaly zone’? Degnan & Rosenberg (2009) defined the anomaly zone as internodes where
there is a higher probability of observing an anomalous
gene tree than the actual species tree. This is caused by the
consecutive arrangement of very short internodes and leads
to symmetric anomalies in asymmetric species trees
(Rosenberg 2013). Suh et al. (2015) suggested that some
neoavian relationships may lie within the anomaly zone.
However, a recent phylogenomic study on skinks emphasized that an accurate species tree topology is essential to
empirically detect the anomaly zone (Linkem et al. 2016).
Considering that there is no bifurcating species tree in a
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A hard polytomy at the root of Neoaves
hard polytomy, the polytomy zone and anomaly zone
should thus be mutually exclusive.
Is the polytomy zone visible in dated species trees of
Neoaves? If this were the case, its near-simultaneous speciation events should translate into extremely short internodes. I tested this by mapping the early neoavian polytomy
zone onto the species tree of Jarvis et al. (2014) and measuring the time from the onset of the neoavian radiation to
the youngest divergence still within in the polytomy zone.
This way of forcing the polytomy zone on a bifurcating
tree suggests that its temporal extent was ~4.618 million
years and internode lengths were similar to those in the
resolvable parts of the neoavian radiation. How is this possible? All these branches exhibit extreme levels of up to
100% ILS (Suh et al. 2015) and thus, as mentioned before,
either Ne must have been large or internode length extremely short (Maddison 1997; Maddison & Knowles 2006).
Now consider the following thought experiment. The average duration from mutation to fixation is 4Ne generations
(Kimura & Ohta 1969). Assuming a generation time of two
years (as present in Ficedula flycatchers, NadachowskaBrzyska et al. 2016), ILS across the putative 4.618 million
years of the neoavian polytomy zone yields an average Ne
of 1 732 352. Although there is evidence for similar Ne in
extant bird species (Nadachowska-Brzyska et al. 2015), it is
hard to imagine that such a large ancestral neoavian population persisted for such a long time and across eight speciation events. Instead, I suggest that the discrepancy
between the neoavian polytomy zone and the relatively
long internodes in the Jarvis et al. (2014) time tree results
from the fact that highly ILS-affected gene trees were
forced on a fully ‘resolved’, yet irresolvable species tree of
Neoaves. Mendes & Hahn (2016) recently showed that an
ILS-related phenomenon, dubbed ‘substitutions produced
by incomplete lineage sorting’ (SPILS), erroneously
increases or decreases internode lengths and thereby affects
a wide range of downstream molecular analyses, including
molecular dates. As mentioned before, the base of Neoaves
witnessed extreme levels of up to 100% ILS across at least
eight speciation events and SPILS can be expected to
strongly affect these internode lengths in any ‘resolved’
species tree. I thus agree that there is a current ‘[i]rrational
exuberance for resolved species trees’ (Hahn & Nakhleh 2016)
and suggest that the internode lengths at the root of
Neoaves should be taken with a grain of salt.
Outlook: what are the implications of hard
polytomies for the animal Tree of Life?
If the early neoavian polytomy is ‘real’, what catalysed the
near-simultaneous emergence of nine bird lineages? Suh
et al. (2015) noted an interesting coincidence between
those branches with extreme degrees of ILS (herein
57
A hard polytomy at the root of Neoaves
A. Suh
collapsed within the polytomy zone; Fig. 2) and the overlap
of their divergence times with the K-Pg boundary in Jarvis
et al. (2014). In the meantime, the Jarvis et al. (2014)’s time
tree was largely confirmed by recent independent analyses
(Claramunt & Cracraft 2015; Prum et al. 2015) whereas
the much older dates estimated via mitochondrial or few
nuclear loci (reviewed, e.g., by Brown & Van Tuinen 2011
and van Tuinen 2009) seem somewhat outdated (but see
Acosta Hospitaleche & Gelfo 2015 and De Pietri et al.
2016). Although it was recently noted that current methods
are unable to confidently distinguish between events right
before and right after the K-Pg boundary (Ksepka & Phillips 2015), these methods imply temporal proximity of the
polytomy zone to the K-Pg boundary. If this assumption is
correct, then it seems plausible that in the wake of the
K-Pg mass extinction of non-avian dinosaurs, archaic birds
and many other organisms, the super-rapid speciation of
Neoaves began with an initial nine-taxon hard polytomy
and subsequently slowed down into a bifurcating tree.
There is another case of super-rapid speciation that
appears to coincide with the K-Pg boundary, namely the
emergence of the three major groups of placental mammals
(O’Leary et al. 2013, but see Springer et al. 2013 and dos
Reis et al. 2014). Various genome-scale studies have
debated whether or not the relationships among Boreotheria, Afrotheria and Xenarthra are resolvable as a bifurcating
tree (Murphy et al. 2007; Churakov et al. 2009; Nishihara
et al. 2009; Hallstr€
om & Janke 2010; Morgan et al. 2013;
Romiguier et al. 2013; Tarver et al. 2016). The most recent
contribution to this controversy has argued that weak phylogenetic signal, not a hard polytomy, is causing the difficulty of resolving the root of placental mammals (Tarver
et al. 2016). Alternatively, only three bifurcating tree
topologies are possible for a three-taxon polytomy of placental mammals (Felsenstein 1978) and I suggest that forcing it on a bifurcating tree might stochastically yield one
topology more often than the other two. But irrespective
of whether or not there exists an early placental mammal
polytomy, this situation is somewhat less complex than the
aforementioned 2 027 025 rooted bifurcating trees which
are possible for a nine-taxon polytomy at the root of
Neoaves. Thus, I hope that the future debate on the
Neoaves problem will be less about which species tree is
‘correct’, and more about if there is such a thing as a fully
bifurcating neoavian Tree of Life at all. In the meantime,
it is likely that we will see the publication of hundreds of
highly discordant species trees that each will be hailed as
the ‘resolved’ one.
In the recent era of phylogenomic research, the early
neoavian polytomy stands out for its complexity. There are
many other genome-scale studies that have resolved substantial phylogenetic conflict in rapid radiations [e.g. bears
58
(Kutschera et al. 2014), Darwin’s finches (Lamichhaney
et al. 2015), Ficedula flycatchers (Nater et al. 2015), gallopheasants (Meiklejohn et al. 2016) and tomatoes (Pease
et al. 2016)]. Notably, none of these cases seem to be irresolvable or contain hard polytomies. However, I predict
that more examples of hard polytomies will be discovered
once various types of densely sampled genome-scale data
are available for other super-rapid radiations. These might
include, for example, gibbons (Veeramah et al. 2015),
Hawaiian honeycreepers (Lerner et al. 2011), Lake Malawi
cichlids (Takahashi et al. 2001), Lake Tanganyika cichlids
(Meyer et al. 2015), Myotis bats (Platt et al. 2015), sylvioid
songbirds (Fregin et al. 2012), thrushes (Nylander et al.
2008) and white eyes (Moyle et al. 2009).
Conclusions
The Neoaves problem pictured in this perspective piece
illustrates the necessity for three paradigm shifts avian phylogenomics as well as genome-scale phylogenomics in general: (i) Bootstrap and conventional Bayesian posterior
probabilities should not be the only measures for topological support. (ii) Every phylogenetic tree hypothesis should
be accompanied by a phylogenetic network for visualization
of conflicts. (iii) Hard polytomies exist in nature and should
be treated as the null hypothesis in the absence of reproducible tree topologies.
In summary, the neoavian Tree of Life is now largely
resolved, except for its deepest relationships which likely
constitute a nine-taxon hard polytomy. One may ask
whether the declaration of a hard polytomy should lead
researchers to turn away from early neoavian relationships
and focus on other issues? On the contrary! Neoaves comprise, to my knowledge, the first empirical example for a
hard polytomy in animals. The imminent genome sequencing of all ~10 500 bird species (Jarvis 2016) will provide an
exciting system for studying the exact phylogenetic extent
and genomic signatures of a hard polytomy.
Acknowledgements
I thank Lutz Bachmann, Per G. P. Ericson and Per Sundberg for the invitation to write this perspective piece. Further thanks go to Per Alstr€
om, David Č ern
y, Matthew W.
Hahn, Erich D. Jarvis, John McCormack, Frank E.
Rheindt, Alexandre P. Selvatti and members of Hans Ellegren’s lab and Jochen Wolf’s lab for numerous helpful discussions. I am grateful to Reto Burri, David Č ern
y, Verena
E. Kutschera, Frank E. Rheindt and Alexandre P. Selvatti
for valuable comments on this manuscript. Further thanks
go to the users of the online discussion platforms Twitter
and Reddit Science for inspiring conversations on the
topics of this manuscript. I am indebted to Jon Fjelds
a for
the permission to use his bird paintings.
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A. Suh
References
Acosta Hospitaleche, C. & Gelfo, J. N. (2015). New Antarctic
findings of Upper Cretaceous and lower Eocene loons (Aves:
Gaviiformes). Annales de Paleontologie, 101, 315–324.
Avise, J. C. & Robinson, T. J. (2008). Hemiplasy: a new term in
the lexicon of phylogenetics. Systematic Biology, 57, 503–507.
Bapteste, E., van Iersel, L., Janke, A., Kelchner, S., Kelk, S.,
McInerney, J. O., Morrison, D. A., Nakhleh, L., Steel, M., Stougie, L. & Whitfield, J. (2013). Networks: expanding evolutionary
thinking. Trends in Genetics, 29, 439–441.
Boc, A., Diallo, A. B. & Makarenkov, V. (2012). T-REX: a web
server for inferring, validating and visualizing phylogenetic trees
and networks. Nucleic Acids Research, 40, W573–W579.
Brown, J. W. & Van Tuinen, M.. 2011. Evolving perceptions on
the antiquity of the modern avian tree. In G. Dyke & G. Kaiser
(Eds) Living Dinosaurs: The Evolutionary History of Modern Birds
(pp. 306–324). Chichester: John Wiley & Sons, Ltd.
Cavalli-Sforza, L. L. & Edwards, A. W. F. (1967). Phylogenetic
analysis Models and estimation procedures. The American Journal
of Human Genetics, 19, 233–257.
Chojnowski, J. L., Kimball, R. T. & Braun, E. L. (2008). Introns
outperform exons in analyses of basal avian phylogeny using clathrin heavy chain genes. Gene, 410, 89–96.
Churakov, G., Kriegs, J. O., Baertsch, R., Zemann, A., Brosius, J.
& Schmitz, J. (2009). Mosaic retroposon insertion patterns in
placental mammals. Genome Research, 19, 868–875.
Claramunt, S. & Cracraft, J. (2015). A new time tree reveals Earth
history’s imprint on the evolution of modern birds. Science
Advances, 1, e1501005.
De Pietri, V. L., Scofield, R. P., Zelenkov, N., Boles, W. E. &
Worthy, T. H. (2016). The unexpected survival of an ancient
lineage of anseriform birds into the Neogene of Australia: the
youngest record of Presbyornithidae. Royal Society Open Science,
3, 150635.
Degnan, J. H. & Rosenberg, N. A. (2009). Gene tree discordance,
phylogenetic inference and the multispecies coalescent. Trends in
Ecology & Evolution, 24, 332–340.
Durand, E. Y., Patterson, N., Reich, D. & Slatkin, M. (2011).
Testing for ancient admixture between closely related populations. Molecular Biology and Evolution, 28, 2239–2252.
Ericson, P. G. P., Anderson, C. L., Britton, T., Elzanowski, A.,
Johansson, U. S., K€allersj€
o, M., Ohlson, J. I., Parsons, T. J.,
Zuccon, D. & Mayr, G. (2006). Diversification of Neoaves: integration of molecular sequence data and fossils. Biology Letters, 2,
543–547.
Fain, M. G. & Houde, P. (2004). Parallel radiations in the primary
clades of birds. Evolution, 58, 2558–2573.
Felsenstein, J. (1978). The number of evolutionary trees. Systematic
Biology, 27, 27–33.
Fregin, S., Haase, M., Olsson, U. & Alstr€
om, P. (2012). New
insights into family relationships within the avian superfamily
Sylvioidea (Passeriformes) based on seven molecular markers.
BMC Evolutionary Biology, 12, 1–12.
Gibb, G. C., Kardailsky, O., Kimball, R. T., Braun, E. L. &
Penny, D. (2007). Mitochondrial genomes and avian phylogeny:
complex characters and resolvability without explosive radiations.
Molecular Biology and Evolution, 24, 269–280.
Hackett, S. J., Kimball, R. T., Reddy, S., Bowie, R. C. K.,
Braun, E. L., Braun, M. J., Chojnowski, J. L., Cox, W. A.,
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A hard polytomy at the root of Neoaves
Han, K.-L., Harshman, J., Huddleston, C. J., Marks, B. D.,
Miglia, K. J., Moore, W. S., Sheldon, F. H., Steadman, D.
W., Witt, C. C. & Yuri, T. (2008). A phylogenomic study of
birds reveals their evolutionary history. Science, 320, 1763–
1768.
Hahn, M. W. & Nakhleh, L. (2016). Irrational exuberance for
resolved species trees. Evolution, 70, 7–17.
Hallstr€
om, B. M. & Janke, A. (2010). Mammalian evolution may
not be strictly bifurcating. Molecular Biology and Evolution, 27,
2804–2816.
Han, K.-L., Braun, E. L., Kimball, R. T., Reddy, S., Bowie, R. C.
K., Braun, M. J., Chojnowski, J. L., Cox, W. A., Han, K.-L.,
Harshman, J., Huddleston, C. J., Marks, B. D., Miglia, K. J.,
Moore, W. S., Sheldon, F. H., Steadman, D. W., Witt, C. C. &
Yuri, T. (2011). Are transposable element insertions homoplasy
free?: an examination using the avian tree of life. Systematic Biology, 60, 375–386.
Harrison, R. G. & Larson, E. L. (2014). Hybridization, introgression, and the nature of species boundaries. Journal of Heredity,
105, 795–809.
Hoelzer, G. A. & Meinick, D. J. (1994). Patterns of speciation and
limits to phylogenetic resolution. Trends in Ecology & Evolution,
9, 104–107.
Huson, D. H. & Bryant, D. (2006). Application of phylogenetic
networks in evolutionary studies. Molecular Biology and Evolution,
23, 254–267.
Jarvis, E. D. (2016). Perspectives from the Avian Phylogenomics
Project: questions that can be answered with sequencing all genomes of a vertebrate class. Annual Review of Animal Biosciences, 4,
45–59.
Jarvis, E. D., Mirarab, S., Aberer, A. J., Li, B., Houde, P., Li, C.,
Ho, S. Y. W., Faircloth, B. C., Nabholz, B., Howard, J. T., Suh,
A., Weber, C. C., da Fonseca, R. R., Li, J., Zhang, F., Li, H.,
Zhou, L., Narula, N., Liu, L., Ganapathy, G., Boussau, B., Bayzid, M. S., Zavidovych, V., Subramanian, S., Gabald
on, T.,
Capella-Gutierrez, S., Huerta-Cepas, J., Rekepalli, B., Munch,
K., Schierup, M., Lindow, B., Warren, W. C., Ray, D., Green,
R. E., Bruford, M. W., Zhan, X., Dixon, A., Li, S., Li, N.,
Huang, Y., Derryberry, E. P., Bertelsen, M. F., Sheldon, F. H.,
Brumfield, R. T., Mello, C. V., Lovell, P. V., Wirthlin, M., Schneider, M. P. C., Prosdocimi, F., Samaniego, J. A., Velazquez,
~ ez, A., Campos, P. F., Petersen, B., SicherA. M. V., Alfaro-N
un
itz-Ponten, T., Pas, A., Bailey, T., Scofield, P., Bunce, M., Lambert, D. M., Zhou, Q., Perelman, P., Driskell, A. C., Shapiro,
B., Xiong, Z., Zeng, Y., Liu, S., Li, Z., Liu, B., Wu, K., Xiao, J.,
Yinqi, X., Zheng, Q., Zhang, Y., Yang, H., Wang, J., Smeds, L.,
Rheindt, F. E., Braun, M., Fjeldsa, J., Orlando, L., Barker, F.
K., Jønsson, K. A., Johnson, W., Koepfli, K.-P., O’Brien, S.,
Haussler, D., Ryder, O. A., Rahbek, C., Willerslev, E., Graves,
G. R., Glenn, T. C., McCormack, J., Burt, D., Ellegren, H.,
Alstr€
om, P., Edwards, S. V., Stamatakis, A., Mindell, D. P., Cracraft, J., Braun, E. L., Warnow, T., Jun, W., Gilbert, M. T. P.
& Zhang, G. (2014). Whole genome analyses resolve the early
branches in the tree of life of modern birds. Science, 346, 1320–
1331.
Jarvis, E. D., Mirarab, S., Aberer, A. J., Li, B., Houde, P., Li, C.,
Ho, S. Y. W., Faircloth, B. C., Nabholz, B., Howard, J. T., Suh,
~ez, A., Narula,
A., Weber, C. C., da Fonseca, R. R., Alfaro-N
un
N., Liu, Burt, L., D., Ellegren, H., Edwards, S. V., Stamatakis,
59
A hard polytomy at the root of Neoaves
A. Suh
A., Mindell, D. P., Cracraft, J., Braun, E. L., Warnow, T., Jun,
W., Gilbert, M. T. P. & Zhang, G. (2015). Phylogenomic analyses data of the Avian Phylogenomics Project. GigaScience, 4, 4.
Jeffroy, O., Brinkmann, H., Delsuc, F. & Philippe, H. (2006). Phylogenomics: the beginning of incongruence? Trends in Genetics,
22, 225–231.
Kimura, M. & Ohta, T. (1969). The average number of generations until fixation of a mutant gene in a finite population.
Genetics, 61, 763–771.
Ksepka, D. T. & Phillips, M. J. (2015). Avian diversification patterns across the K-Pg boundary: influence of calibrations, datasets, and model misspecification. Annals of the Missouri Botanical
Garden, 100, 300–328.
Kuritzin, A., Kischka, T., Schmitz, J. & Churakov, G. (2016).
Incomplete lineage sorting and hybridization statistics for largescale retroposon insertion data. PLoS Computational Biology, 12,
e1004812.
Kutschera, V. E., Bidon, T., Hailer, F., Rodi, J. L., Fain, S. R. &
Janke, A. (2014). Bears in a forest of gene trees: Phylogenetic
inference is complicated by incomplete lineage sorting and gene
flow. Molecular Biology and Evolution, 31, 2004–2017.
Lamichhaney, S., Berglund, J., Almen, M. S., Maqbool, K., Grabherr, M., Martinez-Barrio, A., Promerova, M., Rubin, C.-J.,
Wang, C., Zamani, N., Grant, B. R., Grant, P. R., Webster, M.
T. & Andersson, L. (2015). Evolution of Darwin’s finches and
their beaks revealed by genome sequencing. Nature, 518, 371–375.
Lerner, H. R. L., Meyer, M., James, H. F., Hofreiter, M. & Fleischer, R. C. (2011). Multilocus resolution of phylogeny and timescale in the extant adaptive radiation of Hawaiian honeycreepers.
Current Biology, 21, 1838–1844.
Lewis, P. O., Holder, M. T. & Holsinger, K. E. (2005). Polytomies and Bayesian phylogenetic inference. Systematic Biology,
54, 241–253.
Linkem, C. W., Minin, V. N. & Leache, A. D. (2016). Detecting
the anomaly zone in species trees and evidence for a misleading
signal in higher-level skink phylogeny (Squamata: Scincidae).
Systematic Biology, 65, 465–477.
Liu, L., Xi, Z., Wu, S., Davis, C. C. & Edwards, S. V. (2015).
Estimating phylogenetic trees from genome-scale data. Annals of
the New York Academy of Sciences, 1360, 36–53.
Livezey, B. C. & Zusi, R. L. (2007). Higher-order phylogeny of
modern birds (Theropoda, Aves: Neornithes) based on comparative anatomy II. Analysis and discussion. Zoological Journal of the
Linnean Society, 149, 1–95.
Maddison, W. P. (1997). Gene trees in species trees. Systematic
Biology, 46, 523–536.
Maddison, W. P. & Knowles, L. L. (2006). Inferring phylogeny
despite incomplete lineage sorting. Systematic Biology, 55, 21–30.
Martin, S. H., Davey, J. W. & Jiggins, C. D. (2015). Evaluating
the use of ABBA–BABA statistics to locate introgressed loci.
Molecular Biology and Evolution, 32, 244–257.
Mayr, G. (2011). Metaves, Mirandornithes, Strisores and other
novelties – a critical review of the higher-level phylogeny of
neornithine birds. Journal of Zoological Systematics and Evolutionary Research, 49, 58–76.
McCormack, J. E., Harvey, M. G., Faircloth, B. C., Crawford, N.
G., Glenn, T. C. & Brumfield, R. T. (2013). A phylogeny of
birds based on over 1,500 loci collected by target enrichment
and high-throughput sequencing. PLoS ONE, 8, e54848.
60
Meiklejohn, K. A., Faircloth, B. C., Glenn, T. C., Kimball, R. T.
& Braun, E. L. (2016). Analysis of a rapid evolutionary radiation
using ultraconserved elements (UCEs): Evidence for a bias in
some multispecies coalescent methods. Systematic Biology, 65,
612–627.
Mendes, F. K. & Hahn, M. W. (2016). Gene tree discordance
causes apparent substitution rate variation. Systematic Biology, 65,
711–721.
Meyer, B. S., Matschiner, M. & Salzburger, W. (2015). A tribal
level phylogeny of Lake Tanganyika cichlid fishes based on a
genomic multi-marker approach. Molecular Phylogenetics and Evolution, 83, 56–71.
Morgan, C. C., Foster, P. G., Webb, A. E., Pisani, D., McInerney,
J. O. & O’Connell, M. J. (2013). Heterogeneous models place
the root of the placental mammal phylogeny. Molecular Biology
and Evolution, 30, 2145–2156.
Moyle, R. G., Filardi, C. E., Smith, C. E. & Diamond, J. (2009).
Explosive Pleistocene diversification and hemispheric expansion
of a “great speciator”. Proceedings of the National Academy of
Sciences, 106, 1863–1868.
Murphy, W. J., Pringle, T. H., Crider, T. A., Springer, M. S. &
Miller, W. (2007). Using genomic data to unravel the root of
the placental mammal phylogeny. Genome Research, 17, 413–421.
Nadachowska-Brzyska, K., Li, C., Smeds, L., Zhang, G. & Ellegren, H. (2015). Temporal dynamics of avian populations during
Pleistocene revealed by whole-genome sequences. Current Biology, 25, 1375–1380.
Nadachowska-Brzyska, K., Burri, R., Smeds, L. & Ellegren, H.
(2016). PSMC analysis of effective population sizes in molecular
ecology and its application to black-and-white Ficedula flycatchers. Molecular Ecology, 25, 1058–1072.
Nater, A., Burri, R., Kawakami, T., Smeds, L. & Ellegren, H.
(2015). Resolving evolutionary relationships in closely related
species with whole-genome sequencing data. Systematic Biology,
64, 1000–1017.
Nishihara, H., Maruyama, S. & Okada, N. (2009). Retroposon
analysis and recent geological data suggest near-simultaneous
divergence of the three superorders of mammals. Proceedings of
the National Academy of Sciences, 106, 5235–5240.
Nylander, J. A. A., Olsson, U., Alstr€
om, P. & Sanmartın, I. (2008).
Accounting for phylogenetic uncertainty in biogeography: A
Bayesian approach to dispersal-vicariance analysis of the thrushes
(Aves: Turdus). Systematic Biology, 57, 257–268.
O’Leary, M. A., Bloch, J. I., Flynn, J. J., Gaudin, T. J., Giallombardo, A., Giannini, N. P., Goldberg, S. L., Kraatz, B. P.,
Luo, Z.-X., Meng, J., Ni, X., Novacek, M. J., Perini, F. A.,
Randall, Z. S., Rougier, G. W., Sargis, E. J., Silcox, M. T.,
Simmons, N. B., Spaulding, M., Velazco, P. M., Weksler, M.,
Wible, J. R. & Cirranello, A. L. (2013). The placental mammal ancestor and the post–K-Pg radiation of placentals.
Science, 339, 662–667.
Pacheco, M. A., Battistuzzi, F. U., Lentino, M., Aguilar, R. F.,
Kumar, S. & Escalante, A. A. (2011). Evolution of modern
birds revealed by mitogenomics: timing the radiation and origin of major orders. Molecular Biology and Evolution, 28, 1927–
1942.
Pease, J. B. & Hahn, M. W. (2015). Detection and polarization of
introgression in a five-taxon phylogeny. Systematic Biology, 64,
651–662.
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A. Suh
Pease, J. B., Haak, D. C., Hahn, M. W. & Moyle, L. C. (2016).
Phylogenomics reveals three sources of adaptive variation during
a rapid radiation. PLoS Biology, 14, e1002379.
Philippe, H., Brinkmann, H., Lavrov, D. V., Littlewood, D. T. J.,
Manuel, M., W€
orheide, G., & Baurain, D. (2011). Resolving difficult phylogenetic questions: Why more sequences are not
enough. PLoS Biology, 9, e1000602.
Platt, R. N., Zhang, Y., Witherspoon, D. J., Xing, J., Suh, A.,
Keith, M. S., Jorde, L. B., Stevens, R. D. & Ray, D. A. (2015).
Targeted capture of phylogenetically informative Ves SINE
insertions in genus Myotis. Genome Biology and Evolution, 7,
1664–1675.
Poe, S. & Chubb, A. L. (2004). Birds in a bush: five genes indicate
explosive evolution of avian orders. Evolution, 58, 404–415.
Pratt, R. C., Gibb, G. C., Morgan-Richards, M., Phillips, M. J.,
Hendy, M. D. & Penny, D. (2009). Toward resolving deep
Neoaves phylogeny: data, signal enhancement, and priors. Molecular Biology and Evolution, 26, 313–326.
Prum, R. O., Berv, J. S., Dornburg, A., Field, D. J., Townsend, J.
P., Lemmon, E. M., & Lemmon, A. R. (2015). A comprehensive
phylogeny of birds (Aves) using targeted next-generation DNA
sequencing. Nature, 526, 569–573.
Ray, D. A., Xing, J., Salem, A.-H. & Batzer, M. A. (2006). SINEs
of a nearly perfect character. Systematic Biology, 55, 928–935.
dos Reis, M., Donoghue, P. C. J. & Yang, Z. (2014). Neither phylogenomic nor palaeontological data support a Palaeogene origin
of placental mammals. Biology Letters, 10, 20131003.
Rivera-Rivera, C. J. & Montoya-Burgos, J. I. (2016). LS3: A
method for improving phylogenomic inferences when evolutionary rates are heterogeneous among taxa. Molecular Biology and
Evolution, 33, 1625–1634.
Romiguier, J., Ranwez, V., Delsuc, F., Galtier, N. & Douzery, E.
J. P. (2013). Less is more in mammalian phylogenomics: ATrich genes minimize tree conflicts and unravel the root of placental mammals. Molecular Biology and Evolution, 30, 2134–2144.
Rosenberg, N. A. (2013). Discordance of species trees with their
most likely gene trees: a unifying principle. Molecular Biology and
Evolution, 30, 2709–2713.
Salichos, L. & Rokas, A. (2013). Inferring ancient divergences
requires genes with strong phylogenetic signals. Nature, 497,
327–331.
Salichos, L., Stamatakis, A. & Rokas, A. (2014). Novel information
theory-based measures for quantifying incongruence among
phylogenetic trees. Molecular Biology and Evolution, 31, 1261–1271.
Sangster, G. (2005). A name for the flamingo–grebe clade. Ibis,
147, 612–615.
Seehausen, O. (2004). Hybridization and adaptive radiation. Trends
in Ecology & Evolution, 19, 198–207.
Shedlock, A. M., Takahashi, K. & Okada, N. (2004). SINEs of
speciation: tracking lineages with retroposons. Trends in Ecology
& Evolution, 19, 545–553.
Sibley, C. G. & Ahlquist, J. E. (1990). Phylogeny and Classification of Birds: A Study in Molecular Evolution. New Haven: Yale
University Press.
Slowinski, J. B. (2001). Molecular polytomies. Molecular Phylogenetics and Evolution, 19, 114–120.
Solıs-Lemus, C. & Ane, C. (2016). Inferring phylogenetic networks with maximum pseudolikelihood under incomplete lineage
sorting. PLoS Genetics, 12, e1005896.
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62
A hard polytomy at the root of Neoaves
Springer, M. S., Meredith, R. W., Teeling, E. C. & Murphy, W.
J. (2013). Technical comment on “The placental mammal ancestor and the post–K-Pg radiation of placentals”. Science, 341,
613–613.
Suh, A., Paus, M., Kiefmann, M., Churakov, G., Franke, F. A.,
Brosius, J., Kriegs, J. O. & Schmitz, J. (2011). Mesozoic retroposons reveal parrots as the closest living relatives of passerine
birds. Nature Communications, 2, 443.
Suh, A., Kriegs, J. O., Donnellan, S., Brosius, J. & Schmitz, J.
(2012). A universal method for the study of CR1 retroposons in
nonmodel bird genomes. Molecular Biology and Evolution, 29,
2899–2903.
Suh, A., Smeds, L. & Ellegren, H. (2015). The dynamics of
incomplete lineage sorting across the ancient adaptive radiation
of neoavian birds. PLoS Biology, 13, e1002224.
Takahashi, K., Nishida, M., Yuma, M. & Okada, N. (2001).
Retroposition of the AFC family of SINEs (Short
Interspersed Repetitive Elements) before and during the adaptive radiation of cichlid fishes in Lake Malawi and related
inferences about phylogeny. Journal of Molecular Evolution, 53,
496–507.
Tarver, J. E., dos Reis, M., Mirarab, S., Moran, R. J., Parker, S.,
O’Reilly, J. E., King, B. L., O’Connell, M. J., Asher, R. J., Warnow, T., Peterson, K. J., Donoghue, P. C. J. & Pisani, D.
(2016). The interrelationships of placental mammals and the
limits of phylogenetic inference. Genome Biology and Evolution, 8,
330–344.
vanTuinen, M. (2009). Advanced birds (Neoaves). In S. B. Hedges
& S. Kumar (Eds) The Timetree of Life (pp. 419–422). New York:
Oxford University Press.
van Tuinen, M., Butvill, D. B., Kirsch, J. A. W. & Hedges, S. B.
(2001). Convergence and divergence in the evolution of aquatic
birds. Proceedings of the Royal Society B, 268, 1345–19 1350.
Veeramah, K. R., Woerner, A. E., Johnstone, L., Gut, I., Gut, M.,
Marques-Bonet, T., Carbone, L., Wall, J. D. & Hammer, M. F.
(2015). Examining phylogenetic relationships among gibbon
genera using whole genome sequence data using an approximate
Bayesian computation approach. Genetics, 200, 295–308.
Walsh, H. E., Kidd, M. G., Moum, T. & Friesen, V. L. (1999).
Polytomies and the power of phylogenetic inference. Evolution,
53, 932–937.
Wang, N., Braun, E. L. & Kimball, R. T. (2012). Testing
hypotheses about the sister group of the Passeriformes using an
independent 30-locus dataset. Molecular Biology and Evolution, 29,
737–750.
Whitfield, J. B. & Lockhart, P. J. (2007). Deciphering ancient
rapid radiations. Trends in Ecology & Evolution, 22, 258–265.
Yang, Z. (2007). Fair-balance paradox, star-tree paradox, and Bayesian phylogenetics. Molecular Biology and Evolution, 24, 1639–
1655.
Yang, Z. & Rannala, B. (2005). Branch-length prior influences
Bayesian posterior probability of phylogeny. Systematic Biology,
54, 455–470.
Yu, Y., Barnett, R. M. & Nakhleh, L. (2013). Parsimonious inference of hybridization in the presence of incomplete lineage sorting. Systematic Biology, 62, 738–751.
Yu, Y., Dong, J., Liu, K. J. & Nakhleh, L. (2014). Maximum likelihood inference of reticulate evolutionary histories. Proceedings of
the National Academy of Sciences, 111, 16448–16453.
61
A hard polytomy at the root of Neoaves
A. Suh
Yuri, T., Kimball, R. T., Harshman, J., Bowie, R. C. K., Braun,
M. J., Chojnowski, J. L., Han, K.-L., Hackett, S. J., Huddleston,
C. J., Moore, W. S., Reddy, S., Sheldon, F. H., Steadman, D.
W., Witt, C. C. & Braun, E. L. (2013). Parsimony and modelbased analyses of indels in avian nuclear genes reveal congruent
and incongruent phylogenetic signals. Biology, 2, 419–444.
Zhang, G., Li, C., Li, Q., Li, B., Larkin, D. M., Lee, C., Storz, J.
€
F., Antunes, A., Greenwold, M. J., Meredith, R. W., Odeen,
A.,
Cui, J., Zhou, Q., Xu, L., Pan, H., Wang, Z., Jin, L., Zhang, P.,
Hu, H., Yang, W., Hu, J., Xiao, J., Yang, Z., Liu, Y., Xie, Q.,
Yu, H., Lian, J., Wen, P., Zhang, F., Li, H., Zeng, Y., Xiong,
Z., Liu, S., Zhou, L., Huang, Z., An, N., Wang, J., Zheng, Q.,
Xiong, Y., Wang, G., Wang, B., Wang, J., Fan, Y., da Fonseca,
~ez, A., Schubert, M., Orlando, L., Mourier,
R. R., Alfaro-N
un
T., Howard, J. T., Ganapathy, G., Pfenning, A., Whitney, O.,
Rivas, M. V., Hara, E., Smith, J., Farre, M., Narayan, J., Slavov,
G., Romanov, M. N., Borges, R., Machado, J. P., Khan, I.,
Springer, M. S., Gatesy, J., Hoffmann, F. G., Opazo, J. C.,
H
astad, O., Sawyer, R. H., Kim, H., Kim, K.-W., Kim, H. J.,
Cho, S., Li, N., Huang, Y., Bruford, M. W., Zhan, X., Dixon,
A., Bertelsen, M. F., Derryberry, E., Warren, W., Wilson, R.
K., Li, S., Ray, D. A., Green, R. E., O’Brien, S. J., Griffin, D.,
Johnson, W. E., Haussler, D., Ryder, O. A., Willerslev, E.,
Graves, G. R., Alstr€
om, P., Fjelds
a, J., Mindell, D. P., Edwards,
62
S. V., Braun, E. L., Rahbek, C., Burt, D. W., Houde, P., Zhang,
Y., Yang, H., Wang, J., Avian Genome Consortium, Jarvis, E.
D., Gilbert, M. T. P. & Wang, J. (2014). Comparative genomics
reveals insights into avian genome evolution and adaptation.
Science, 346, 1311–1320.
Supporting Information
Additional Supporting Information may be found in the
online version of this article:
Fig. S1. Topological discordance of the eight first neoavian branching events in the main trees of (A) Jarvis et al.
(2014) and (B) Prum et al. (2015).
Fig. S2. (A) Majority-rule consensus tree based on Fig. 2
and (B) simulated nine-taxon polytomy based on a sample
of 100% of the 512 possible retroposon presence/absence
patterns. Colour coding of phylogenetic networks corresponds to Fig. 2.
Fig. S3. Consensus networks of (A) 3,679 UCE gene
trees and (B) 2,022 binned intron gene trees from Jarvis
et al. (2014).
ª 2016 Royal Swedish Academy of Sciences, 45, s1, October 2016, pp 50–62