Next Article in Journal
Venomous Snake Abundance Within Snake Species’ Assemblages Worldwide
Previous Article in Journal
Effects of Forest Fragmentation on the Vertical Stratification of Neotropical Bats
Open AccessArticle

Phylogenomic Reconstruction Sheds Light on New Relationships and Timescale of Rails (Aves: Rallidae) Evolution

1
Molecular Epidemiology and Public Health Laboratory, Hopkirk Research Institute, School of Veterinary Science, Massey University, Private Bag, 11 222, Palmerston North 4442, New Zealand
2
Department of Biological Science, Florida State University, 319 Stadium Drive, PO Box 3064295, Tallahassee, FL 32306-4295, USA
3
Department of Scientific Computing, Florida State University, Dirac Science Library, Tallahassee, FL 32306-4102, USA
*
Author to whom correspondence should be addressed.
Diversity 2020, 12(2), 70; https://doi.org/10.3390/d12020070
Received: 21 October 2019 / Revised: 28 January 2020 / Accepted: 4 February 2020 / Published: 7 February 2020

Abstract

The integration of state-of-the-art molecular techniques and analyses, together with a broad taxonomic sampling, can provide new insights into bird interrelationships and divergence. Despite their evolutionary significance, the relationships among several rail lineages remain unresolved as does the general timescale of rail evolution. Here, we disentangle the deep phylogenetic structure of rails using anchored phylogenomics. We analysed a set of 393 loci from 63 species, representing approximately 40% of the extant familial diversity. Our phylogenomic analyses reconstruct the phylogeny of rails and robustly infer several previously contentious relationships. Concatenated maximum likelihood and coalescent species-tree approaches recover identical topologies with strong node support. The results are concordant with previous phylogenetic studies using small DNA datasets, but they also supply an additional resolution. Our dating analysis provides contrasting divergence times using fossils and Bayesian and non-Bayesian approaches. Our study refines the evolutionary history of rails, offering a foundation for future evolutionary studies of birds.
Keywords: evolution; phylogenomics; Rallidae; systematics; timetree evolution; phylogenomics; Rallidae; systematics; timetree

1. Introduction

The rails (Aves: Rallidae) are a remarkable group of birds with many secretive species that are difficult to observe and others that have managed to colonise very isolated islands despite their apparent poor ability to fly [1,2,3,4]. It is a widespread family of birds extensively distributed throughout insular and continental settings and only absent in polar regions, waterless deserts, and mountains above the snow line. Interest in rail diversity, evolution, and biogeography has led to studies of their relationships using morphology and PCR-based Sanger approaches [5,6]. However, phylogenetic inferences for rails have proven difficult with morphology [5], and despite considerable effort with molecules (e.g., [6,7,8,9,10]), their relationships and diversification dynamics remain poorly understood, with disagreements regarding key relationships due to few markers and species availability. Recently, one study utilized mitogenomes to explore their basal relationships and diversification, but sampling was too limited for understanding deep evolutionary patterns as they lacked good taxon representation at the genus level [11].
Garcia-R et al. [12] published the most comprehensive molecular phylogeny of the rails using approximately 70% of recognised extant and recently extinct rail species diversity. This study clarified many of the systematic issues within the group and established eight main clades (“Aramides”, “Rallus”, “Fulica”, “Laterallus”, “Porzana”, “Gallicrex”, Porphyrio, and Rallina) to help resolve taxonomic confusion arising from evolutionary convergence. Several important aspects of the systematics and taxonomy of the rails have been in flux after the study of Garcia-R et al. [12], including the re-establishment of the genus Zapornia to several species previously placed in Porzana and the description of a new genus in recent years [13]. Furthermore, some relationships are just beginning to settle, including the monotypic ocellated crake (Micropygia schomburgkii) from the tropical forest of the New World [14] and a species within Atlantisia [15]. Nonetheless, some relationships remain largely unresolved or understudied, despite their evolutionary significance, and the general timescale of rail evolution is poorly known. For example, the relationships of the endemic genus Rallicula from New Guinea and the monotypic Rouget’s rail (Rougetius rougetii) from east Africa have not been explored from a molecular perspective, and their placement on the rail phylogeny is uncertain.
The deep origin of the group is also a matter of debate, because important discrepancies persist in the phylogenetic signals retrieved from fossils and DNA data [16,17,18,19,20,21,22,23]. The adequate diagnostic material of fossil rails in continental deposits and subfossil insular endemics has been assigned to modern genera, showing that young crown group lineages were present during Pliocene and Pleistocene times (<5 Mya) [24,25], while older fossils appear not to be directly ancestral to current rallid crown groups [26,27,28,29]. Garcia-R et al. [11], using mitochondrial DNA (mtDNA) genomes, showed that the temporal origin and diversification of the rails occurred during the Eocene ca. 40.5 (49–33) Mya. By comparison, previous studies with reduced sampling using nuclear DNA (nDNA) sequences and different calibration constraints [17,30,31] estimated the origin of the Rallidae to be of the Miocene age ca. 20 Mya, nearly half the age estimated for mtDNA.
Genome-scale data can facilitate the construction of well-supported phylogenies across the Tree of Life and improve the estimation of accurate and precise molecular dating. The recent application of dense taxon sampling using large numbers of genes through modern phylogenomic approaches has resolved relationships and dates in groups of birds (e.g., [22]) and other vertebrates (e.g., [32]). Here, we disentangle the deep phylogenetic structure of the rails using anchored phylogenomics. This is the first evolutionary hypothesis of the rails derived from genomic-scale data, and we aim to use these data to resolve several previously contentious relationships. We analysed a dataset of 393 loci for 63 species, representing approximately 40% of the extant familial diversity, including all major lineages and deep branches of the tree. The taxa selected were chosen attempting to maximize sampling from different clades while prioritising time and resources. The integration of state-of-the-art molecular techniques, together with a broad taxonomic sampling and analytical approaches, provides new insights into rail relationships and evolution.

2. Materials and Methods

2.1. Data Collection

We collected tissues from 63 rails, including species not previously found in other molecular-based approaches. Genomic DNA was extracted using the standard protocol of the QIAGEN Dneasy kit. Prior to library preparation, the quantity and quality of the DNA extractions were inspected using Qubit and a 2% TAE agarose gel, respectively. Sequencing data were generated in an Illumina HiSeq2500 platform at the Center for Anchored Phylogenenomics at Florida State University (www.anchoredphylogeny.com), following Lemmon et al. [33]. Briefly, the DNA extracted was fragmented (150–350 bp) using a Covaris E220 focused-ultrasonicator with Covaris microTUBES. Subsequently, library preparation and indexing were performed on a Beckman-Coulter Biomek FXp liquid-handling robot, following a protocol modified from Meyer and Kircher [34]. Anchored hybrid enrichment was performed using a custom SureSelect kit (Agilent Technologies) targeting loci derived from Prum et al. [22]. Sequencing was performed in the Translational Science Laboratory in the College of Medicine at Florida State University.

2.2. Sequence Assembly and Alignment

After sequencing, paired reads were processed downstream, following the methods described in Prum et al. [22]. Paired reads were compared using all possible degrees of overlap. For each overlap possibility, the number of matches between the two reads was counted. Then, using a binomial model, the probability of obtaining that many matches by chance was computed. Lastly, paired reads were merged when one of these probabilities (corresponding to a particular degree of overlap) was less than 1 × 10−10, and the next smallest probability (corresponding to a different degree of overlap) was more than 1 × 10−7. This stringent filter prevents reads containing repetitive regions from being merged [35]. After reads were merged, mismatches were reconciled using base-specific quality scores, which were combined to form the new quality scores for the merged read [35]. Reads failing to meet the probability criterion were kept separate but still used in the assembly [35]. Reads were assembled into contigs using a pipeline described by Ruane et al. [36] and Hamilton et al. [37]. After filtering out consensus sequences generated from fewer than 100 reads, sets of orthologous sequences were obtained based on pairwise sequence distances. Sequences were aligned using MAFFT v7.023b [38] with “– genafpair – maxiterate 1000” flags. The alignment for each locus was then trimmed/masked, following Ruane et al. [36] and Hamilton et al. [37], with “good” sites identified as those containing >40% identity and fewer than 12 missing/unmasked characters removed from the alignment. The data are available in the Zenodo archive (http://10.5281/zenodo.3576641).

2.3. Phylogenetic Analyses and Divergence Time Estimates

The sequences from similar loci to those included in our dataset were extracted and aligned from other birds using data provided by Prum et al. [22] (Supplementary Table S1). The interpretation of the phylogenetic relationships was based primarily on this full dataset with concatenated and coalescent species tree approaches in RAxML v8.2.8 [39] and ASTRAL-II v4.10.12 [40], respectively. The maximum likelihood (ML) trees were estimated for each locus individually and for the concatenated alignment partitioned by the gene using a GTR + gamma model and a rapid bootstrap with 1000 replicates [41]. A coalescent species tree was summarized with ASTRAL-II using individual trees as inputs and multi-locus bootstrapping. ASTRAL-II uses a heuristic search to find the species tree that agrees with the largest number of quartet trees induced by the set of input gene trees.
We used calibration constraints from the fossil information of Gruiformes and Rallidae to estimate the divergence times in BEAST2 v2.4.7 [42]. We applied the method of Claramunt and Cracraft [23], in which prior calibration densities were modeled based on available fossil records. We placed an exponential prior on the crown ages of Gruiformes (offset: 52, mean: 8.5) and a lognormal prior on the age of the Rallidae (offset: 32.6, mean: 1.1, standard deviation: 1.8) based on densities inferred by Stervander et al. [15], which were themselves based on thoroughly examined fossil data [43,44]. Belgirallus is considered a well-constrained and the oldest fossil record of the Rallidae [27,28], but it is sometimes shown outside the crown group Rallidae [26]. We then also used Belgirallus as a stem group representative to compare and assess the impact on ages predicted. We combined the results of two independent runs of 50 million generations where chains were sampled every 5000th generation using a Birth–Death process prior. A burn-in of 20% was implemented to obtain ESS values above 200. Given the computational limits, the Bayesian coalescent species tree method in BEAST2 was applied to a subset of the data. To reduce the computational time demanded from the large dataset, we randomly selected a subset of 34 loci (a total of 57,241 nucleotides) with the most complete data for all taxa and variable sites > 25% (to maximize the coverage of sites) [45]. The substitution model for each locus was determined in jModelTest2 [46] using the Bayesian information criteria. For comparison, we performed a timetree analysis using a non-Bayesian model in RelTime [47]. This approach implemented in MEGAX [48] allows for using the full anchored loci dataset without the computer burn caused by BEAST2. We used the GTR + Γ model with five gamma categories, fixed the topology to that of the ML tree, and implemented the same calibration parameters as in BEAST2. For this analysis, we used an ML tree that included a more extensive set of bird clades as an outgroup (Supplementary Table S1).

3. Results

3.1. Phylogenetics

Our dataset includes 393 anchored loci and 669,861 sites, 268,317 of which were variables for the dataset comprising the rails (63 taxa) and other birds (Supplementary Table S1). Both the concatenated and coalescent trees produced congruent tree topologies (Figure 1 and Figure S1). These analyses reconstructed the Forbes’s Forest rail (Rallicula forbesi) separate from the rails and sister to the flufftail (Sarothrura rufa), rendering the Rallidae as non-monophyletic. The previous clades assigned in Garcia-R et al. [12] were mostly consistent, but the deep relationships among those clades differed and are best supported in the phylogenomics analyses. The interrelationships within the clades are extensively congruent with the results in Garcia-R et al. [12]. However, the African Nkulengu rail (Himantornis haematopus) was now found in its own separate clade as previously suggested [5,49,50]. The tree inferences using our anchored dataset revealed the phylogenetic positions of some taxa previously not studied. New species not included in Garcia-R et al. [12] were the slaty-breasted wood rail (Aramides saracura), which has relationships within the “Aramides” clade; the Inaccessible rail (Atlantisia rogersi), the ocellated crake (Micropygia schomburgkii), and the red-and-white crake (Laterallus leucopyrrhus) found in “Laterallus”; the African crake (Crex egregia) and the Rouget’s rail (Rougetius rougetii) in “Rallus”; and the white-browed crake (Amaurornis cinerea) and the striped crake (Amaurornis marginalis) within the “Gallicrex” clade.

3.2. Molecular Dating

The divergence times based on Belgirallus as a stem and crown group representative estimated the origin of the rails in the Late Oligocene around 26 (21–32) Mya (Figure 2A) and the Early Oligocene around 33 (32–36) Mya (Figure 2B), respectively, using the Bayesian method in BEAST2. Several early lineage splitting events in Rallidae occurred during the Oligocene, and the major clades originated in the Miocene. The alternative non-Bayesian approach in RelTime provided slightly similar divergence time estimates than those in BEAST2 when Belgirallus was part of crown Rallidae but younger when placed as part of Ralloidea. The time for the origin of Rallidae obtained in RelTime (Figure S2) using the topology of the ML tree and the complete concatenated multi-locus dataset was around 19 (16–23) Mya and 34 Mya (32–44) for the crown Ralloidea and Rallidae calibrations, respectively.

4. Discussion

We reported the relationship of the rails using a 393-gene dataset with dense taxonomic sampling (63 rail species). The evolutionary relationships in our analyses are mostly in agreement with those found using multi-gene phylogenetic datasets [12]. However, our dataset outperforms previous concatenated alignments in terms of data quantity and provides strong statistical support for deep clade divergences, except for the splitting of Himantornis + “Fulica”, and Porphyrio, which receives <90% bootstrap support. The relationships of species within Rallicula are resolved now, showing that they are outside of Rallidae and sister to Sarothrura [12] and Mentocrex [14], and are all better treated as the distinct family Sarothruridae. Both Rallicula and Sarothrura present sexual dichromatism, a condition only present in two species of the rails, the watercock (Gallicrex cinerea) and the little crake (Zapornia parva) [49]. Another novel result includes the proposal of a distinctive and separate placement of the African Nkulengu rail, an outcome concordant with morphological data [5,49] but conflicting with previous molecular analysis [12]. The apparent explanations for this could be the use of an incorrectly identified sample in Garcia-R et al. [12] or the increased molecular data in our study that unambiguously place this species separate from other clades. The genus Crex, composed of two species, the corn crake (C. crex) and the African crake (C. egregia), is not monophyletic. The African crake is shown in our phylogenetic analysis to form a monophyletic group with the Rouget’s rail (Rougetius rougetii), also found in Africa. Our analysis confirms the relationship among the ocellated crake (Micropygia schomburgkii) and russet-crowned crake (Rufirallus viridis) as suggested in a previous molecular phylogenetic study [14]. The enigmatic white-browed crake (Amaurornis cinerea), which has been placed in several genera, including Poliolimnas, Porphyrio, and Porzana, is now shown in our analysis to be closely related to the New Guinea flightless rail (Megacrex inepta). This result, and the clustering of the watercock (Gallicrex cinerea), renders Amaurornis as a non-monophyletic group despite the current placing of several species, previously assigned to this genus, to Zapornia (e.g., Z. akool and Z. flavirostra). Our results also showed that the Inaccessible rail (Atlantisia rogersi) is embedded within the “Laterallus” clade and closely related to the black rail (Laterallus jamaicensis). The current generic assignment of this species to Atlantisia will show Laterallus as non-monophyletic. The taxonomic transferring of the Inaccessible rail to Laterallus has been recommended by Stervander et al. [15], which will also include the Ascension crake (A. elpenor), if its placement to the monotypic Mundia [51] is unjustified. Testing a hypothesis of the relationship of these species will be needed using a dataset that includes the Ascension crake. Future taxonomic changes are warranted for species within Crex, Gallirallus, Porzana, and Gallinula [13].
The unresolved affinities of Belgirallus and other Oligocene rail-like fossils [26,29] make it difficult to estimate the approximate age of crown Rallidae. These fossils are only known from partial limb bones, and there are no current phylogenies that place them into the crown Rallidae. Furthermore, identification of these fossils is complicated by the paraphyly of traditional Rallidae, and some of the earlier rallid fossils seem to belong to Sarothruridae [52]. Our crown and stem inferences, nonetheless, place an uncertainty on Belgirallus and provide a range of calibration points for divergence time analyses. The divergence times obtained using a Bayesian relaxed-clock model and a non-Bayesian model were mostly similar when Belgirallus was used as a crown Rallidae representative. As illustrated in our node dating analyses, the root and origin of Rallidae were estimated to be 33 (32–36) Mya using a Bayesian approach or 34 (32–44) Mya with a non-Bayesian method. However, the credibility intervals (uncertainty) for the origin of Rallidae estimated in RelTime were rather wide compared with the 95% highest posterior density (HPD) intervals in BEAST2. RelTime has been used in studies with large datasets [53], which are intractable with Bayesian analysis, showing an efficient way to estimate divergence times [54]. Previous inferences of divergence times using RelTime have been in disagreement with Bayesian relaxed molecular clocks, [55] but this seems to be related to the priors assigned in Bayesian approaches [56]. The divergence times obtained from both methods when Belgirallus was used as a crown representative of Ralloidea were significantly younger, especially with RelTime, and in concordance with earlier studies that dated the clade’s crown group between 18 (12–24) and 22 (13–36) Mya [17,30,31,57]. Most recent studies have recovered significantly older ages around 40.5 (49–33) Mya [11,14] that conflict markedly with conclusive rail fossils identified in the Early Miocene [58]. The difficulty in estimating a consistent divergence times of the Rallidae crown group originates in part from the scarcity of reliable old crown fossils, the calibration constraints used, the methods implemented, and the relationship between branch lengths from mitochondrial versus nuclear data [18,59,60,61]. The uncertainty in the timescale of rails can be refined and updated as new early fossils are discovered.
Large-scale, multilocus data, combined with improved analytical tools for inferring phylogenetic trees and new methods and fossils for tree calibration, provide unprecedented opportunities for resolving phylogenetic relationships and estimating the divergence time of monophyletic groups. The rails are frequently undersampled in phylogenomic studies (e.g., [22,62]), and our results provide a robust framework for building the phylogenomic supertree of birds. This is important for the study of avian macroevolution and biogeography [63] and for resolving discrepancies in the phylogenetic signals retrieved from geo-temporal fossils and molecular data.

Supplementary Materials

The following are available online at https://www.mdpi.com/1424-2818/12/2/70/s1; Figure S1: Complete phylogenetic relationships among the rails and other birds based on the coalescent ASTRAL analysis; Figure S2: Timetrees estimated with RelTime with Belgirallus as a stem (A) and crown (B) group representative. The bar in each node represents confidence intervals.; Table S1: List of species of the rails and other birds from Prum et al. [22] used in this study. The Handbook of the Birds of the World (HBW) was followed for current nomenclature and species delimitations. Acronyms for museums are the following: AMNH = American Museum of Natural History, USA; ANWC = Australian National Wildlife Collection, Australia; EBU = Evolutionary Biology Unit at the Australian Museum, Australia; FMNH = Field Museum of Natural History, USA; KU = University of Kansas Division of Ornithology, USA; LSUMZ = Louisiana State University Museum of Zoology, USA; NHMO = Natural History Museum University of Oslo, Norway; UAM = University of Alaska Museum, USA; UH = University of Heidelberg; USNM = National Museum of Natural History, Smithsonian Institution, USA; MZUSP = Museu de Zoologia, Universidade de São Paulo, Brazil; UWBM = University of Washington Burke Museum, USA; YPM = Peabody Museum of Natural History, Yale University, USA; and ZMUC = Zoological Museum, University of Copenhagen, Denmark. An asterisk (*) indicates additional species used in the RelTime analysis.

Author Contributions

Conceptualization, J.C.G.-R.; methodology, J.C.G.-R., E.M.L., and A.R.L.; software, J.C.G.-R. and A.R.L.; validation, J.C.G.-R., E.M.L., A.R.L., and N.F.; formal analysis, J.C.G.-R. and A.R.L.; investigation, J.C.G.-R.; resources, J.C.G.-R. and A.R.L.; data curation, J.C.G.-R.; writing—original draft preparation, J.C.G.-R.; writing—review and editing, J.C.G.-R., E.M.L., A.R.L., and N.F.; visualization, J.C.G.-R.; project administration, J.C.G.-R. and N.F.; and funding acquisition, J.C.G.-R. and N.F. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Acknowledgments

We are grateful to the following individuals and institutions that kindly provided tissues necessary for this study: Gerhard Niko (Germany); Michael Wink at University of Heidelberg (Germany); Lisa Nupen and Ross Wanless at the Percy FitzPatrick Institute of African Ornithology (South Africa); Walter Boles and Jaynia Sladek at the Australian Museum (Australia); Leo Joseph and Robert Palmer at the Australian National Wildlife Collection (Australia); Robb Brumfield and Donna Dittmann at the Louisiana University Natural Museum (USA); Ben Marks at the Field Museum of Natural History (USA); Paul Sweet and Chris Filardi at the American Museum of Natural History (USA); Jan Kristensen at the Zoological Museum University of Copenhagen (Denmark); Luis Fabio Silveira at the Museu de Zoologia da Universidade de São Paulo (Brazil); Kristof Zyskowski at the Peabody Museum of Natural History Yale University (USA); Sharon Birks at the Burke Museum (USA); Kevin Winker and Jack Withrow at University of Alaska Museum (USA); Helen F. James, Christopher Milensky, and Christina A. Gebhard at the Smithsonian Institution (USA); Mark Robbins at Kansas University Natural History Museum (USA); and Lars Johannessen and Arild Johnsen at Natural History Museum University of Oslo (Norway). We thank Michelle Kortyna, Chris Zdyrski, Sean Holland, and Jesse Cherry in the Center for Anchored Phylogenomics for assistance with data collection and analysis. We are grateful to Gerald Mayr, Martin Stervander, and four anonymous reviewers for providing helpful comments that greatly improved this manuscript. Images courtesy of Lip Kee Yap, Ramon Moller Jensen, Tom Tarrant, Trevor Collins, Nigel Voaden, Janos Olah, Herve Michel, and Herman Weerman. Special thanks to The Macaulay Library at the Cornell Lab of Ornithology where the following photos were used: ML128818131, ML23661851, ML174510471, ML30971721, and ML177451401.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Garcia-R, J.C.; Elliott, G.; Walker, K.; Castro, I.; Trewick, S.A. Trans-equatorial range of a land bird lineage (Aves: Rallidae) from tropical forests to subantarctic grasslands. J. Avian Biol. 2016, 47, 219–226. [Google Scholar] [CrossRef]
  2. Garcia-R, J.C.; Joseph, L.; Adcock, G.; Reid, J.; Trewick, S.A. Interisland gene flow among populations of the buff-banded rail (Aves: Rallidae) and its implications for insular endemism in Oceania. J. Avian Biol. 2017, 48, 679–690. [Google Scholar] [CrossRef]
  3. Garcia-R, J.C.; Trewick, S.A. Dispersal and speciation in purple swamphens (Rallidae: Porphyrio). The Auk 2015, 132, 140–155. [Google Scholar] [CrossRef]
  4. Garcia-R, J.C.; Gonzalez-Orozco, C.E.; Trewick, S.A. Contrasting patterns of diversification in a bird family (Aves: Gruiformes: Rallidae) are revealed by analysis of geospatial distribution of species and phylogenetic diversity. Ecography 2019, 42, 500–510. [Google Scholar] [CrossRef]
  5. Livezey, B.C. A phylogenetic analysis of the Gruiformes (Aves) based on morphological characters, with an emphasis on the rails (Rallidae). Philos. Trans. R. Soc. B-Biol. Sci. 1998, 353, 2077–2151. [Google Scholar] [CrossRef]
  6. Slikas, B.; Olson, S.L.; Fleischer, R.C. Rapid, independent evolution of flightlessness in four species of Pacific Island rails (Rallidae): An analysis based on mitochondrial sequence data. J. Avian Biol. 2002, 33, 5–14. [Google Scholar] [CrossRef]
  7. Ruan, L.; Wang, Y.; Hu, J.; Ouyang, Y. Polyphyletic origin of the genus Amaurornis inferred from molecular phylogenetic analysis of rails. Biochem. Genet. 2012, 50, 959–966. [Google Scholar] [CrossRef]
  8. Chen, P.; Han, Y.; Zhu, C.; Gao, B.; Ruan, L. Complete mitochondrial genome of Porzana fusca and Porzana pusilla and phylogenetic relationship of 16 Rallidae species. Genetica 2017, 145, 559–573. [Google Scholar] [CrossRef]
  9. Gong, J.; Zhao, R.; Huang, Q.; Sun, X.; Huang, L.; Jing, M. Two mitogenomes in Gruiformes (Amaurornis akool/A. phoenicurus) and the phylogenetic placement of Rallidae. Genes Genom. 2017, 39, 987–995. [Google Scholar] [CrossRef]
  10. Kirchman, J.J. Speciation of flightless rails on islands: A DNA-based phylogeny of the typical rails of the Pacific. The Auk 2012, 129, 56–69. [Google Scholar]
  11. Garcia-R, J.C.; Gibb, G.C.; Trewick, S.A. Eocene diversification of crown group rails (Aves: Gruiformes: Rallidae). PLoS ONE 2014, 9, e109635. [Google Scholar] [CrossRef] [PubMed]
  12. Garcia-R, J.C.; Gibb, G.C.; Trewick, S.A. Deep global evolutionary radiation in birds: Diversification and trait evolution in the cosmopolitan bird family Rallidae. Mol. Phylogenet. Evol. 2014, 81, 96–108. [Google Scholar] [CrossRef] [PubMed]
  13. Sangster, G.; Garcia-R, J.C.; Trewick, S.A. A new genus for the Lesser Moorhen Gallinula angulata Sundevall, 1850 (Aves, Rallidae). Eur. J. Taxon. 2015, 153, 1–8. [Google Scholar] [CrossRef]
  14. Boast, A.P.; Chapman, B.; Herrera, M.B.; Worthy, T.H.; Scofield, R.P.; Tennyson, A.J.D.; Houde, P.; Bunce, M.; Cooper, A.; Mitchell, K.J. Mitochondrial Genomes from New Zealand’s Extinct Adzebills (Aves: Aptornithidae: Aptornis) Support a Sister-Taxon Relationship with the Afro-Madagascan Sarothruridae. Diversity 2019, 11, 24. [Google Scholar] [CrossRef]
  15. Stervander, M.; Ryan, P.G.; Melo, M.; Hansson, B. The origin of the world’s smallest flightless bird, the Inaccessible Island Rail Atlantisia rogersi (Aves: Rallidae). Mol. Phylogenet. Evol. 2019, 130, 92–98. [Google Scholar] [CrossRef]
  16. Ho, S.Y.W.; Duchêne, S. Molecular-clock methods for estimating evolutionary rates and timescales. Mol. Ecol. 2014, 23, 5947–5965. [Google Scholar] [CrossRef]
  17. Brown, J.W.; Payne, R.B.; Mindell, D.P. Nuclear DNA does not reconcile ‘rocks’ and ‘clocks’ in Neoaves: A comment on Ericson et al. Biol. Lett. 2007, 3, 257–260. [Google Scholar] [CrossRef]
  18. Ksepka, D.T.; Ware, J.L.; Lamm, K.S. Flying rocks and flying clocks: disparity in fossil and molecular dates for birds. P. Roy. Soc. B: Bio. Sci. 2014, 281. [Google Scholar] [CrossRef]
  19. Cranston, K.; Rannala, B. Molecular clocks: Closing the gap between rocks and clocks. Heredity 2005, 94, 461–462. [Google Scholar] [CrossRef]
  20. Cracraft, J.; Houde, P.; Ho, S.Y.W.; Mindell, D.P.; Fjeldså, J.; Lindow, B.; Edwards, S.V.; Rahbek, C.; Mirarab, S.; Warnow, T.; et al. Response to Comment on “Whole-genome analyses resolve early branches in the tree of life of modern birds”. Science 2015, 349, 1460. [Google Scholar] [CrossRef]
  21. Mitchell, K.J.; Cooper, A.; Phillips, M.J. Comment on “Whole-genome analyses resolve early branches in the tree of life of modern birds”. Science 2015, 349, 1460. [Google Scholar] [CrossRef] [PubMed]
  22. Prum, R.O.; Berv, J.S.; Dornburg, A.; Field, D.J.; Townsend, J.P.; Lemmon, E.M.; Lemmon, A.R. A comprehensive phylogeny of birds (Aves) using targeted next-generation DNA sequencing. Nature 2015, 526, 569–573. [Google Scholar] [CrossRef] [PubMed]
  23. Claramunt, S.; Cracraft, J. A new time tree reveals Earth history’s imprint on the evolution of modern birds. Sci. Adv. 2015, 1, e1501005. [Google Scholar] [CrossRef] [PubMed]
  24. Steadman, D.W. Extinction and Biogeography of Tropical Pacific Birds; The University of Chicago Press: Chicago, IL, USA, 2006. [Google Scholar]
  25. Hume, J.P.; Walters, M. Extinct Birds; Bloomsbury: London, UK, 2012. [Google Scholar]
  26. De Pietri, V.L.; Mayr, G. Reappraisal of early Miocene rails (Aves, Rallidae) from central France: diversity and character evolution. J. Zool. Syst. Evol. Res. 2014, 52, 312–322. [Google Scholar] [CrossRef]
  27. Mayr, G. A rail (Aves, Rallidae) from the early Oligocene of Germany. Ardea 2006, 94, 23–31. [Google Scholar]
  28. Mayr, G.; Smith, R. Ducks, rails, and limicoline waders (Aves: Anseriformes, Gruiformes, Charadriiformes) from the lowermost Oligocene of Belgium. Geobios 2001, 34, 547–561. [Google Scholar] [CrossRef]
  29. Mayr, G.; Bochenski, Z.M. A skeleton of a small rail from the Rupelian of Poland adds to the diversity of early Oligocene Ralloidea. Neues Jahrb. Geol. Paläontol. - Abhandlungen 2016, 282, 125–134. [Google Scholar] [CrossRef]
  30. Ericson, P.G.P.; Anderson, C.L.; Britton, T.; Elzanowski, A.; Johansson, U.S.; Kallersjo, M.; Ohlson, J.I.; Parsons, T.J.; Zuccon, D.; Mayr, G. Diversification of Neoaves: Integration of molecular sequence data and fossils. Biol. Lett. 2006, 2, 543–547. [Google Scholar] [CrossRef]
  31. Fain, M.G.; Krajewski, C.; Houde, P. Phylogeny of ‘‘core Gruiformes’’ (Aves: Grues) and resolution of the Limpkin–Sungrebe problem. Mol. Phylogenet. Evol. 2007, 43, 515–529. [Google Scholar] [CrossRef]
  32. Chen, X.; Lemmon, A.R.; Lemmon, E.M.; Alexander Pyron, R.; Burbrink, F.T. Using phylogenomics to understand the link between biogeographic origins and regional diversification in ratsnakes. Mol. Phylogenet. Evol. 2017, 111, 206–218. [Google Scholar] [CrossRef]
  33. Lemmon, A.; Emme, S.; Lemmon, E. Anchored hybrid enrichment for massively high-throughput phylogenomics. Syst. Biol. 2012, 61, 727. [Google Scholar] [CrossRef] [PubMed]
  34. Meyer, M.; Kircher, M. Illumina sequencing library preparation for highly multiplexed target capture and sequencing. Cold Spring Harb. Protoc. 2010, 2010, pdb.prot5448. [Google Scholar] [CrossRef] [PubMed]
  35. Rokyta, D.R.; Lemmon, A.R.; Margres, M.J.; Aronow, K. The venom-gland transcriptome of the eastern diamondback rattlesnake (Crotalus adamanteus). BMC Genom. 2012, 13, 312. [Google Scholar] [CrossRef]
  36. Ruane, S.; Raxworthy, C.J.; Lemmon, A.R.; Lemmon, E.M.; Burbrink, F.T. Comparing species tree estimation with large anchored phylogenomic and small Sanger-sequenced molecular datasets: an empirical study on Malagasy pseudoxyrhophiine snakes. BMC Evol. Biol. 2015, 15, 221. [Google Scholar] [CrossRef] [PubMed]
  37. Hamilton, C.A.; Lemmon, A.R.; Lemmon, E.M.; Bond, J.E. Expanding anchored hybrid enrichment to resolve both deep and shallow relationships within the spider tree of life. BMC Evol. Biol. 2016, 16, 212. [Google Scholar] [CrossRef] [PubMed]
  38. Katoh, K.; Standley, D.M. MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Mol. Biol. Evol. 2013, 30, 772–780. [Google Scholar] [CrossRef] [PubMed]
  39. Stamatakis, A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 2014, 30, 1312–1313. [Google Scholar] [CrossRef]
  40. Mirarab, S.; Warnow, T. ASTRAL-II: Coalescent-based species tree estimation with many hundreds of taxa and thousands of genes. Bioinformatics 2015, 31, i44–i52. [Google Scholar] [CrossRef]
  41. Abadi, S.; Azouri, D.; Pupko, T.; Mayrose, I. Model selection may not be a mandatory step for phylogeny reconstruction. Nat. Commun. 2019, 10, 934. [Google Scholar] [CrossRef]
  42. Bouckaert, R.; Heled, J.; Kühnert, D.; Vaughan, T.; Wu, C.-H.; Xie, D.; Suchard, M.A.; Rambaut, A.; Drummond, A.J. BEAST 2: A Software Platform for Bayesian Evolutionary Analysis. PLoS Comput. Biol. 2014, 10, e1003537. [Google Scholar] [CrossRef]
  43. Musser, G.; Ksepka, D.T.; Field, D.J. New Material of Paleocene-Eocene Pellornis (Aves: Gruiformes) Clarifies the Pattern and Timing of the Extant Gruiform Radiation. Diversity 2019, 11, 102. [Google Scholar] [CrossRef]
  44. Bertelli, S.; Chiappe, L.M.; Mayr, G. A new Messel rail from the Early Eocene Fur Formation of Denmark (Aves, Messelornithidae). J. Syst. Palaeontol. 2011, 9, 551–562. [Google Scholar] [CrossRef]
  45. Marin, J.; Hedges, S.B. Undersampling Genomes has Biased Time and Rate Estimates Throughout the Tree of Life. Mol. Biol. Evol. 2018, msy103. [Google Scholar] [CrossRef] [PubMed]
  46. Darriba, D.; Taboada, G.L.; Doallo, R.; Posada, D. jModelTest 2: more models, new heuristics and parallel computing. Nat. Methods 2012, 9. [Google Scholar] [CrossRef] [PubMed]
  47. Tamura, K.; Battistuzzi, F.U.; Billing-Ross, P.; Murillo, O.; Filipski, A.; Kumar, S. Estimating divergence times in large molecular phylogenies. Proc. Natl. Acad. Sci. USA 2012, 109, 19333–19338. [Google Scholar] [CrossRef]
  48. Kumar, S.; Stecher, G.; Li, M.; Knyaz, C.; Tamura, K. MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Mol. Biol. Evol. 2018, 35, 1547–1549. [Google Scholar] [CrossRef]
  49. Olson, S.L. A classification of the Rallidae. Wilson Bull. 1973, 85, 381–416. [Google Scholar]
  50. Ripley, S.D. Rails of the world: A monograph of the Family Rallidae; David, R., Ed.; Godine Publisher: Boston, MA, USA, 1977. [Google Scholar]
  51. Bourne, W.R.P.; Ashmole, N.P.; Simmons, K.E.L. A New Subfossil Night Heron and a New Genus for the Extinct Rail from Ascension Island, Central Tropical Atlantic Ocean. Ardea 2003, 91, 45–51. [Google Scholar]
  52. Mayr, G. Avian Evolution: The Fossil Record of Birds and its Paleobiological Significance; Wiley: West Sussex, UK, 2017. [Google Scholar]
  53. Irisarri, I.; Singh, P.; Koblmüller, S.; Torres-Dowdall, J.; Henning, F.; Franchini, P.; Fischer, C.; Lemmon, A.R.; Lemmon, E.M.; Thallinger, G.G.; et al. Phylogenomics uncovers early hybridization and adaptive loci shaping the radiation of Lake Tanganyika cichlid fishes. Nat. Commun. 2018, 9, 3159. [Google Scholar] [CrossRef]
  54. Tamura, K.; Tao, Q.; Kumar, S. Theoretical Foundation of the RelTime Method for Estimating Divergence Times from Variable Evolutionary Rates. Mol. Biol. Evol. 2018, 35, 1770–1782. [Google Scholar] [CrossRef]
  55. Lozano-Fernandez, J.; dos Reis, M.; Donoghue, P.C.J.; Pisani, D. RelTime Rates Collapse to a Strict Clock When Estimating the Timeline of Animal Diversification. Genome Biol. Evol. 2017, 9, 1320–1328. [Google Scholar] [CrossRef] [PubMed]
  56. Battistuzzi, F.U.; Tao, Q.; Jones, L.; Tamura, K.; Kumar, S. RelTime Relaxes the Strict Molecular Clock throughout the Phylogeny. Genome Biol. Evol. 2018, 10, 1631–1636. [Google Scholar] [CrossRef] [PubMed]
  57. Ericson, P.G.P.; Anderson, C.L.; Mayr, G. Hangin’on to our rocks’n clocks: a reply to Brown et al. Biol. Lett. 2007, 3, 260–261. [Google Scholar] [CrossRef]
  58. Mather, E.K.; Tennyson, A.J.D.; Scofield, R.P.; De Pietri, V.L.; Hand, S.J.; Archer, M.; Handley, W.D.; Worthy, T.H. Flightless rails (Aves: Rallidae) from the early Miocene St Bathans Fauna, Otago, New Zealand. J. Syst. Palaeontol. 2018, 1–27. [Google Scholar] [CrossRef]
  59. Lukoschek, V.; Scott Keogh, J.; Avise, J.C. Evaluating Fossil Calibrations for Dating Phylogenies in Light of Rates of Molecular Evolution: A Comparison of Three Approaches. Syst. Biol. 2011, 61, 22. [Google Scholar] [CrossRef]
  60. Sauquet, H.; Ho, S.Y.W.; Gandolfo, M.A.; Jordan, G.J.; Wilf, P.; Cantrill, D.J.; Bayly, M.J.; Bromham, L.; Brown, G.K.; Carpenter, R.J.; et al. Testing the Impact of Calibration on Molecular Divergence Times Using a Fossil-Rich Group: The Case of Nothofagus (Fagales). Syst. Biol. 2011, 61, 289–313. [Google Scholar] [CrossRef]
  61. Marshall, C.R. Using the Fossil Record to Evaluate Timetree Timescales. Front. Genet. 2019, 10, 1049. [Google Scholar] [CrossRef]
  62. Reddy, S.; Kimball, R.T.; Pandey, A.; Hosner, P.A.; Braun, M.J.; Hackett, S.J.; Han, K.-L.; Harshman, J.; Huddleston, C.J.; Kingston, S.; et al. Why Do Phylogenomic Data Sets Yield Conflicting Trees? Data Type Influences the Avian Tree of Life more than Taxon Sampling. Syst. Biol. 2017, 66, 857–879. [Google Scholar] [CrossRef]
  63. Kimball, R.T.; Oliveros, C.H.; Wang, N.; White, N.D.; Barker, F.K.; Field, D.J.; Ksepka, D.T.; Chesser, R.T.; Moyle, R.G.; Braun, M.J.; et al. A Phylogenomic Supertree of Birds. Diversity 2019, 11, 109. [Google Scholar] [CrossRef]
Figure 1. RAxML phylogeny of the rails and other Gruiformes based on a 393-gene matrix. Bootstrap support values are shown at nodes. Coloured boxes indicate clades as in Garcia-R et al. [12] where brown = “Aramides”; yellow = “Rallus”; black = “Fulica”; blue = Porphyrio; purple = “Laterallus”; green = Rallina; red = “Porzana”; and orange = “Gallicrex”.
Figure 1. RAxML phylogeny of the rails and other Gruiformes based on a 393-gene matrix. Bootstrap support values are shown at nodes. Coloured boxes indicate clades as in Garcia-R et al. [12] where brown = “Aramides”; yellow = “Rallus”; black = “Fulica”; blue = Porphyrio; purple = “Laterallus”; green = Rallina; red = “Porzana”; and orange = “Gallicrex”.
Diversity 12 00070 g001
Figure 2. Chronogram of the rails evolution for the 34-loci dataset based on a relaxed-clock model using the calibration configuration of crown Gruiformes and Belgirallus as a stem (A) and crown (B) group representative. For each node, the estimated time of divergence is indicated with a bar representing the 95% highest posterior density (HPD) intervals. Coloured boxes represent major clades as in Figure 1.
Figure 2. Chronogram of the rails evolution for the 34-loci dataset based on a relaxed-clock model using the calibration configuration of crown Gruiformes and Belgirallus as a stem (A) and crown (B) group representative. For each node, the estimated time of divergence is indicated with a bar representing the 95% highest posterior density (HPD) intervals. Coloured boxes represent major clades as in Figure 1.
Diversity 12 00070 g002
Back to TopTop