Abstract

Whole plastid genomes are being sequenced rapidly from across the green plant tree of life, and phylogenetic analyses of these are increasing resolution and support for relationships that have varied among or been unresolved in earlier single- and multi-gene studies. Pooideae, the cool-season grass lineage, is the largest of the 12 grass subfamilies and includes important temperate cereals, turf grasses and forage species. Although numerous studies of the phylogeny of the subfamily have been undertaken, relationships among some ‘early-diverging’ tribes conflict among studies, and some relationships among subtribes of Poeae have not yet been resolved. To address these issues, we newly sequenced 25 whole plastomes, which showed rearrangements typical of Poaceae. These plastomes represent 9 tribes and 11 subtribes of Pooideae, and were analysed with 20 existing plastomes for the subfamily. Maximum likelihood (ML), maximum parsimony (MP) and Bayesian inference (BI) robustly resolve most deep relationships in the subfamily. Complete plastome data provide increased nodal support compared with protein-coding data alone at nodes that are not maximally supported. Following the divergence of Brachyelytrum, Phaenospermateae, Brylkinieae–Meliceae and Ampelodesmeae–Stipeae are the successive sister groups of the rest of the subfamily. Ampelodesmeae are nested within Stipeae in the plastome trees, consistent with its hybrid origin between a phaenospermatoid and a stipoid grass (the maternal parent). The core Pooideae are strongly supported and include Brachypodieae, a Bromeae–Triticeae clade and Poeae. Within Poeae, a novel sister group relationship between Phalaridinae and Torreyochloinae is found, and the relative branching order of this clade and Aveninae, with respect to an Agrostidinae–Brizinae clade, are discordant between MP and ML/BI trees. Maximum likelihood and Bayesian analyses strongly support Airinae and Holcinae as the successive sister groups of a Dactylidinae–Loliinae clade.

Introduction

Advances in next-generation sequencing technologies (Moore et al. 2006; Cronn et al. 2008; Parks et al. 2009; Wysocki et al. 2014) have resulted in a rapid increase in completed plastid genomes (Jansen and Ruhlman 2012) sampled widely across the green plant tree of life. The use of whole plastomes to infer phylogenies (i.e. phylogenomics) provides opportunities to potentially increase resolution and support for relationships that have varied among or been unresolved and/or poorly supported in earlier single- and multi-gene studies. Plastomes have been used to address diverse phylogenetic questions at deep (Ruhfel et al. 2014) to shallow (Parks et al. 2009) hierarchical levels, and to characterize plastid genome evolution (e.g. patterns of gene loss and organization, GC content, microstructural events, evolutionary rates) (e.g. Barrett and Davis 2012; Jansen and Ruhlman 2012).

Grasses are the fourth largest family of flowering plants in the world, with some 11 000 species and 600–900 genera. At the family level, numerous phylogenetic studies have been conducted and the deep phylogenetic framework for Poaceae is well established. The family contains three small, deeply diverging subfamilies (Anomochlooideae, Pharoideae, Puelioideae) that are the successive sister groups of a large clade comprising two major lineages, the Bambusoideae, Ehrhartoideae, Pooideae (BEP) and the Panicoideae, Arundinoideae, Chloridoideae, Micrairoideae, Aristidoideae, Danthonioideae (PACMAD) clades (Grass Phylogeny Working Group 2001; Duvall et al. 2007; Sanchez-Ken and Clark 2007; Bouchenak-Khelladi et al. 2008; Saarela and Graham 2010; Grass Phylogeny Working Group II 2012).

The BEP clade includes the bamboo (Bambusoideae), rice (Ehrhartoideae) and cool-season (Pooideae) grass subfamilies. The origin of its crown clade has been dated at ∼40–53 to 70.7–72.6 million years ago, depending on which fossil calibration points are used (Christin et al. 2014). Earlier analyses inferred varying relationships among the subfamilies of the BEP clade (Hsiao et al. 1998; Hilu et al. 1999; Zhang 2000; Grass Phylogeny Working Group 2001; Duvall et al. 2007) and a few studies found that Pooideae may be the sister group of the PACMAD clade (Soreng and Davis 1998; Duvall et al. 2007), but there is now consensus from plastid multi-gene analyses that Bambusoideae and Pooideae are sister taxa (Bouchenak-Khelladi et al. 2008; Saarela and Graham 2010; Grass Phylogeny Working Group II 2012; Wu and Ge 2012). This is consistent with some symplesiomorphic bambusoid macro- and micromorphological characters in ‘early-diverging’ lineages within Pooideae that were traditionally included in Bambusoideae (Clark et al. 1995). This topology has been confirmed by a phylogenomic study of 121 orthologous nuclear genes (Zhao et al. 2013) and a species tree inferred from >18 000 maximum likelihood (ML) gene trees (Burleigh et al. 2011), but it was not recovered in a nuclear analysis of 43 putative orthologous cDNA sequences (Peng et al. 2010).

Pooideae, the cool-season grass lineage, is the largest in the BEP clade and the largest of the 12 grass subfamilies, with some 189 genera, 10 nothogenera and 3900 species (Grass Phylogeny Working Group 2001; Clayton et al. 2006 onwards; Soreng et al. 2014). It includes important temperate cereals like wheat (Triticum), barley (Hordeum) and oats (Avena), cool-season turf grasses in the genera Festuca, Poa, Agrostis, Lolium and Schedonorus (Beard 2012), and numerous temperate forage species. Since the establishment of Pooideae in the mid-19th century by Bentham (1861), its circumscription has undergone radical realignment as understanding of the evolutionary history of grasses advanced (reviewed in Soreng and Davis 2000; Soreng et al. 2007). Pooideae, in its current circumscription, was first recognized as a natural group in cladistic studies of morphological data (Kellogg and Campbell 1987), and this hypothesis has been corroborated by numerous molecular phylogenetic studies (Soreng et al. 1990; Davis and Soreng 1993, 2007, 2010; Cummings et al. 1994; Nadot et al. 1994; Clark et al. 1995; Catalán et al. 1997; Hsiao et al. 1998; Soreng and Davis 1998; Hilu et al. 1999; Mathews et al. 2000; Grass Phylogeny Working Group 2001; Duvall et al. 2007; Bouchenak-Khelladi et al. 2008; Grass Phylogeny Working Group II 2012; Blaner et al. 2014 [matK analyses]).

Multiple phylogenetic analyses based on plastid and nuclear ribosomal data have clarified the major evolutionary lineages within Pooideae (e.g. Davis and Soreng 2007; Döring et al. 2007; Soreng et al. 2007; Bouchenak-Khelladi et al. 2008; Schneider et al. 2009). These are recognized as supertribes, tribes and/or subtribes in two recent classifications that differ only in the ranks chosen for some lineages. The classification by Soreng et al. (2014), modified from Soreng et al. (2003, 2007), recognizes 2 supertribes, 14 tribes and 1 subtribe, while that of Röser and colleagues (Döring et al. 2007; Schneider et al. 2009, 2011) recognizes 9 tribes, 9 subtribes and 1 tribal complex (Table 1). Here we follow the classification of Soreng et al. (2014)[see Supporting Information].

Table 1.

Comparison of the Pooideae classifications of Soreng et al. (2014) and Schneider et al. (2009, 2011). Parallel taxa (taxon names in boldface) are equivalent in circumscription but may differ in rank. Numbers in square brackets are the number of genera in each taxon according to Soreng et al. (2014). Schneider et al. (2009, 2011) do not provide a subtribal classification for their Aveneae/Poeae complex.

Soreng et al. (2014) Schneider et al. (2009, 2011) 
 Tribe Brachyelytreae Ohwi [1] Tribe Brachyelytreae 
 Tribe Nardeae 
 Tribe Nardeae W.D.J. Koch. [1]  Subtribe Nardinae Kromb. 
 Tribe Lygeae J. Presl [1]  Subtribe Lygeinae Röser 
 Tribe Phaenospermateae Renvoize & Clayton [3] Tribe Phaenospermateae 
  Subtribe Duthieinae Pilg. ex Potztal [5–6] Tribe Duthieae Röser & J. Schneider 
 Tribe Stipeae 
 Tribe Stipeae Dumort. [∼29]  Subtribe Stipinae Griseb. 
 Tribe Ampelodesmeae Tutin [1]  Subtribe Ampelosdesminae Conert 
 Tribe Meliceae 
 Tribe Meliceae Link ex Endl. [6]  Subtribe Melicinae Fr. 
 Tribe Brylkinieae Tateoka [2]  Subtribe Brylkiniinae Ohwi 
 Tribe Diarrheneae C.S. Campb. [2] Tribe Diarrheneae 
 Tribe Brachypodieae Harz [1] Tribe Brachypodieae 
Supertribe Triticodae T.D. Macfarl. & L. Watson Tribe Hordeae Kunth ex Spenn. 
Littledalea Hemsl. [1]  Subtribe Littledaleinae Röser 
 Tribe Bromeae Dumort. [1]  Subtribe Brominae Dumort. 
 Tribe Triticeae Dumort. [∼25] [this name, published in 1824, has priority over Hordeae Kunth ex Spenn., published in 1825]  Subtribe Hordeinae Dumort. 
Supertribe Poodae L. Liu  
 Tribe Poeae R.Br. [∼120] Aveneae/Poeae complex [unranked] 
 Poeae chloroplast group 1 (Aveneae type) [= Poeae clade 1 in current study]  
  Subtribe Torreyochloinae Soreng  
  Subtribe Aveninae J. Presl  
  Subtribe Phalaridinae Fr.  
  Subtribe Anthoxanthinae A. Gray  
  Subtribe Brizinae Tzvelev s.s.  
  Subtribe Brizinae s.l. “Calotheca Clade”  
  Subtribe Agrostidinae Fr.  
 Poeae chloroplast group 2 (Poeae type) [= Poeae clade 2 in current study]  
  Subtribe Scolochloinae Tzvelev  
  Subtribe Sesleriinae Parl.  
  Subtribe Coleanthinae Rouy  
  Subtribe Miliinae Dumort.  
  Subtribe Poinae Dumort.  
  Subtribe Holcinae Dumort.  
  Subtribe Airinae Fr.  
  Subtribe Loliinae Dumort.  
  Subtribe Dactylidinae Stapf  
  Subtribe Cynosurinae Fr.  
  Subtribe Ammochloinae Tzvelev  
  Subtribe Parapholiinae Caro  
Soreng et al. (2014) Schneider et al. (2009, 2011) 
 Tribe Brachyelytreae Ohwi [1] Tribe Brachyelytreae 
 Tribe Nardeae 
 Tribe Nardeae W.D.J. Koch. [1]  Subtribe Nardinae Kromb. 
 Tribe Lygeae J. Presl [1]  Subtribe Lygeinae Röser 
 Tribe Phaenospermateae Renvoize & Clayton [3] Tribe Phaenospermateae 
  Subtribe Duthieinae Pilg. ex Potztal [5–6] Tribe Duthieae Röser & J. Schneider 
 Tribe Stipeae 
 Tribe Stipeae Dumort. [∼29]  Subtribe Stipinae Griseb. 
 Tribe Ampelodesmeae Tutin [1]  Subtribe Ampelosdesminae Conert 
 Tribe Meliceae 
 Tribe Meliceae Link ex Endl. [6]  Subtribe Melicinae Fr. 
 Tribe Brylkinieae Tateoka [2]  Subtribe Brylkiniinae Ohwi 
 Tribe Diarrheneae C.S. Campb. [2] Tribe Diarrheneae 
 Tribe Brachypodieae Harz [1] Tribe Brachypodieae 
Supertribe Triticodae T.D. Macfarl. & L. Watson Tribe Hordeae Kunth ex Spenn. 
Littledalea Hemsl. [1]  Subtribe Littledaleinae Röser 
 Tribe Bromeae Dumort. [1]  Subtribe Brominae Dumort. 
 Tribe Triticeae Dumort. [∼25] [this name, published in 1824, has priority over Hordeae Kunth ex Spenn., published in 1825]  Subtribe Hordeinae Dumort. 
Supertribe Poodae L. Liu  
 Tribe Poeae R.Br. [∼120] Aveneae/Poeae complex [unranked] 
 Poeae chloroplast group 1 (Aveneae type) [= Poeae clade 1 in current study]  
  Subtribe Torreyochloinae Soreng  
  Subtribe Aveninae J. Presl  
  Subtribe Phalaridinae Fr.  
  Subtribe Anthoxanthinae A. Gray  
  Subtribe Brizinae Tzvelev s.s.  
  Subtribe Brizinae s.l. “Calotheca Clade”  
  Subtribe Agrostidinae Fr.  
 Poeae chloroplast group 2 (Poeae type) [= Poeae clade 2 in current study]  
  Subtribe Scolochloinae Tzvelev  
  Subtribe Sesleriinae Parl.  
  Subtribe Coleanthinae Rouy  
  Subtribe Miliinae Dumort.  
  Subtribe Poinae Dumort.  
  Subtribe Holcinae Dumort.  
  Subtribe Airinae Fr.  
  Subtribe Loliinae Dumort.  
  Subtribe Dactylidinae Stapf  
  Subtribe Cynosurinae Fr.  
  Subtribe Ammochloinae Tzvelev  
  Subtribe Parapholiinae Caro  

Pooideae comprise numerous, mostly [with the exception of Meliceae (150 species) and Stipeae (557); Clayton et al. 2006 onwards] species-poor ‘early-diverging’ lineages (those resulting from the earliest or deepest splits in the clade), and the core Pooideae, which as defined by Soreng et al. (2007), includes the tribes Brachypodieae, Bromeae, Poeae, Triticeae and the genus Littledalea, and includes the majority of species in the subfamily. Most studies identify Brachyelytrum [= Brachyelytreae], a small enigmatic genus that has been variously recognized as pooid or bambusoid (reviewed in Saarela et al. 2003), as the sister group of the rest of the subfamily [Catalán et al. 1997; Grass Phylogeny Working Group 2001; Davis and Soreng 2007 (matK analyses); Duvall et al. 2007; Bouchenak-Khelladi et al. 2008; Davis and Soreng 2010; Grass Phylogeny Working Group II 2012; Blaner et al. 2014], although in a few studies Brachyelytrum falls outside the Pooideae clade [Davis and Soreng 1993; Soreng and Davis 1998; Blaner et al. 2014 (topoisomerase 6 analysis)]. After the divergence of Brachyelytrum, a clade of Lygeum [= Lygeeae] and Nardus [= Nardeae] is the next successive sister group of the rest of the subfamily in multiple studies [Catalán et al. 1997; Soreng and Davis 1998, 2000; Mathews et al. 2000; Grass Phylogeny Working Group 2001; Davis and Soreng 2007, 2010; Döring et al. 2007 (Nardeae not sampled); Duvall et al. 2007; Bouchenak-Khelladi et al. 2008; Schneider et al. 2011; Grass Phylogeny Working Group II 2012; Blaner et al. 2014], with the exception of two in which Brachyelytrum and Nardus form a moderately supported clade that is sister to rest of the subfamily (Hsiao et al. 1998; Hilu and Alice 1999).

The tribes Meliceae, Brylkinieae, Phaenospermateae, Stipeae and Ampelodesmeae are the next emerging branches in the phylogeny. Brylkinieae has been included only in a few analyses, where it has been placed as the sister group of Meliceae (Schneider et al. 2009, 2011; Davis and Soreng 2010; Romaschenko et al. 2010, 2012; Rodionov et al. 2013; Blaner et al. 2014). The monotypic Ampelodesmos (A. mauritanicus) [= Ampelodesmeae] has consistently been nested in Stipeae in plastid and nuclear ribosomal analyses (Hsiao et al. 1998; Davis and Soreng 2007, 2010; Grass Phylogeny Working Group II 2012; Romaschenko et al. 2012), although it differs dramatically in spikelet morphology, being multi-flowered rather than single-flowered like Stipeae. Using a low copy nuclear gene, Romaschenko et al. (2014) recently demonstrated that Ampelodesmos originated via hybridization between a phaenospermatoid and an ‘early-diverging’ stipoid grass (the maternal parent), explaining its placement in Stipeae in plastid trees, and leading Soreng et al. (2014) to accept Ampelodesmeae as a monotypic tribe.

The interrelationships among Stipeae–Ampelodesmeae, Meliceae and Phaenospermateae with respect to the rest of Pooideae have varied substantially among studies based on different taxon samplings, molecular markers and methods of phylogenetic inference. Three studies infer these to be successively diverging lineages with strong support in Bayesian analyses for at least two of the three relevant deep nodes, but in each of these studies the position of at least one of the tribes/lineages differs with respect to the others. The best supported topology is presented in a tree based on two plastid genes, in which Phaenospermateae (not monophyletic in the analysis), Stipeae–Ampelodesmeae and Meliceae are strongly supported as successively diverging lineages (Duvall et al. 2007). In contrast, a tree based on three plastid genes identifies Stipeae–Ampelodesmeae, Phaenospermateae and Meliceae as successively diverging lineages (in other words, the branching order of the first two lineages is reversed), but with weak support for the clade that includes Phaenospermateae, Meliceae and the rest of Pooideae (Grass Phylogeny Working Group II 2012). A nine-region plastid tree supports Phaenospermateae, Brylkinieae–Meliceae and Stipeae–Ampelodesmeae as successively diverging lineages, but with weak support for the clade including Stipeae–Ampelodesmeae and the rest of Pooideae (Romaschenko et al. 2012).

Diarrheneae and Brachypodieae are resolved as the next successively diverging lineages in most analyses, sister to a clade including Bromeae, Littledalea, Poeae and Triticeae [Clark et al. 1995 (Brachypodieae not sampled); Catalán et al. 1997; Hilu et al. 1999 (Diarrheneae not sampled); Grass Phylogeny Working Group 2001; Davis and Soreng 2007; Döring et al. 2007; Duvall et al. 2007 (Bayesian analysis); Schneider et al. 2009 (Diarrheneae not sampled); Davis and Soreng 2010; Schneider et al. 2011 (ITS, but weak support for the Diarrheneae + Brachypodieae + rest of Pooideae clade); Grass Phylogeny Working Group II 2012 (Diarrheneae not sampled)]. In other studies, different topologies were found: the two tribes were resolved as a clade [Bouchenak-Khelladi et al. 2008; Blaner et al. 2014 (matK analysis)], Brachypodieae were inferred to have diverged prior to Diarrheneae (Mathews et al. 2000), the two lineages comprised a polytomy with the rest of the core Pooideae (Schneider et al. 2011) and Diarrheneae were part of an unresolved polytomy with Stipeae, Meliceae, Phaenospermateae, Ampelodesmeae and the core Pooideae [Blaner et al. 2014 (based on sequences of the nuclear gene topoisomerase 6)]. In one ITS analysis, Diarrheneae were weakly supported as the sister group of the rest of Pooideae (Hsiao et al. 1998).

Among the core Pooideae, Triticeae and Brachypodium are estimated to have diverged 32–39 Mya, based on comparisons of their nuclear genomes (The International Brachypodium Initiative 2010); this represents the origin of the crown core Pooideae. An independent estimate places the origin of the Triticeae–Poeae split at ∼26–33.5 Mya (Sandve and Fjellheim 2010). There is consensus that Triticeae and Bromeae are sister taxa (Catalán et al. 1997; Hsiao et al. 1998; Soreng and Davis 1998, 2000; Hilu et al. 1999; Mathews et al. 2000; Grass Phylogeny Working Group 2001; Davis and Soreng 2007, 2010; Döring et al. 2007; Duvall et al. 2007; Bouchenak-Khelladi et al. 2008; Schneider et al. 2009, 2011; Grass Phylogeny Working Group II 2012; Blaner et al. 2014) and all studies that have sampled Littledalea have found this genus to be the sister group of the Bromeae–Triticeae clade (Davis and Soreng 2007, 2010; Döring et al. 2007; Soreng et al. 2007; Schneider et al. 2009, 2011; Blaner et al. 2014), with the exception of one in which cloned topoisomerase 6 sequences of Littledalea tibetica formed a polytomy with the Bromeae–Triticeae clade (Blaner et al. 2014).

Poeae is the most species-rich of the tribes of Pooideae, with some 2258 species distributed in cool-temperate, Mediterranean and arctic climates (Clayton et al. 2006 onwards). Taxa now included in Poeae have been variously arranged in multiple smaller tribes and subtribes (reviewed by Soreng et al. 2007; see also Quintanar et al. 2007; Gillespie et al. 2008). Of these, the Aveneae (the oat tribe) and the Poeae s.s. have been recognized most widely, distinguished on the basis of several morphological characters (e.g. Clayton and Renvoize 1986), some of which are quite homoplasious (Soreng et al. 2007). Multiple studies have demonstrated that Aveneae and Poeae s.s. are not monophyletic in any of their traditional circumscriptions (e.g. Soreng and Davis 2000; Davis and Soreng 2007; Döring et al. 2007; Quintanar et al. 2007; Soreng et al. 2007; Schneider et al. 2009). Within Poeae s.l. two major clades have been identified in analyses of plastid deoxyribonucleic acid (DNA), which have been variously recognized informally (Soreng and Davis 2000; Saarela et al. 2010; Grass Phylogeny Working Group II 2012; Soreng et al. 2014); we refer to these simply as Poeae clades 1 and 2, as in Schneider et al. (2009). Six subtribes are recognized in Poeae clade 1 and 12 subtribes in Poeae clade 2 (Table 1). Evolutionary relationships within and among these lineages have been addressed in numerous studies across the tribe (Soreng et al. 2007; Schneider et al. 2009, 2011, 2012) and within and among tribes: Airinae and Holcinae (Chiapella 2007), Aveninae (Grebenstein et al. 1998; Röser et al. 2001; Rodionov et al. 2005; Winterfeld et al. 2009a, b, 2014; Romero-Zarco 2011), Brizinae (Essi et al. 2008), Loliinae (Torrecilla and Catalán 2002; Catalán et al. 2004; Torrecilla et al. 2004; Inda et al. 2008), Phalaridinae (Voshell et al. 2011), Poinae (Gillespie et al. 2007, 2008, 2009, 2010; Hoffmann et al. 2013), Poeae clade 1 (Quintanar et al. 2007, 2010; Saarela et al. 2010). However, numerous aspects of the relationships within and among the subtribes of Poeae remain unclear.

Grass plastomes

The economically important grasses were among the first organisms to have their plastid genomes sequenced. The plastome of rice (Hiratsuka et al. 1989) was the third species, after tobacco and a liverwort, and the first monocot completed; and the maize plastome (Maier et al. 1995) was the sixth one completed (Jansen et al. 2005). There are now (as of 25 September 2014) some 106 plastid genomes publicly available for grasses, representing 43 genera,78 species and 8 subfamilies: Anomochlooideae (1 species) (Givnish et al. 2010; Morris and Duvall 2010; Jones et al. 2014), Pharoideae (2 species) (Jones et al. 2014), Puelioideae (1 species) (Jones et al. 2014), Bambusoideae (33 species) (Wu et al. 2009; Zhang et al. 2011; Burke et al. 2012, 2014; Wu and Ge 2012; Gao and Gao 2014; Ma et al. 2014), Ehrhartoideae (7 species) (Hiratsuka et al. 1989; Shahid Masood et al. 2004; Tang et al. 2004; Wu and Ge 2012; Lin et al. 2014), Panicoideae (7 species) (Maier et al. 1995; Asano et al. 2004; Calsa Júnior et al. 2004; Saski et al. 2007; Diekmann et al. 2009; Leseberg and Duvall 2009; Young et al. 2011; Besnard et al. 2013), Pooideae (48 species) (Ogihara et al. 2000; Saski et al. 2007; Diekmann et al. 2009; Hand et al. 2013; Gornicki et al. 2014; Middleton et al. 2014) and Chloridoideae (1 species) (Wysocki et al. 2014) [see Supporting Information].

Several studies have generated whole plastomes for grasses, and phylogenetic analyses of these plastomes have resulted in increased resolution and support for relationships within and among some grass subfamilies compared with trees in earlier single- and multi-gene studies. For example, Jones et al. (2014) found the two-genus subfamily Anomochlooideae to be monophyletic, a result found in some (Clark et al. 1995; Grass Phylogeny Working Group 2001; Duvall et al. 2007; Grass Phylogeny Working Group II 2012) but not all [Hilu et al. 1999; Zhang 2000; Bouchenak-Khelladi et al. 2008; Blaner et al. 2014 (matK analyses)] previous analyses, and they reconstructed species-level relationships in Pharoideae. Several plastome studies support a sister group relationship between Bambusoideae and Pooideae (Zhang et al. 2011; Wu and Ge 2012; Burke et al. 2014; Jones et al. 2014—but see the ML tree in Young et al. 2011) and in Bambusoideae plastomes have substantially improved resolution and support within and among species of the temperate woody bamboo tribe Arundinarieae (Zhang et al. 2011; Ma et al. 2014) compared with earlier multi-locus plastid phylogenies (Triplett and Clark 2010; Zeng et al. 2010). In Triticeae, plastomes have clarified relationships among closely related species of wheat (Triticum) and goatgrass (Aegilops) (Gornicki et al. 2014; Middleton et al. 2014). These examples indicate that whole plastomes hold much promise for resolving relationships among grass clades that have previously been problematic. Here, we report 25 new plastomes of taxa of Pooideae and use these in combination with previously published plastomes to infer phylogenetic relationships among the major lineages in the subfamily.

Methods

Taxon sampling and DNA extraction

Silica-dried leaf tissue was obtained from 25 species of pooid grasses (Table 2). Tissue was homogenized manually in liquid nitrogen before extraction. The DNA extraction protocol was followed using the Qiagen DNeasy Plant Mini Kit (Qiagen Inc., Valencia, CA, USA).

Table 2.

Voucher specimen information and GenBank accession numbers for newly sequenced plastomes, place of publication of previously published plastomes and Illumina library preparation methods (TruSeq, Nextera or TruSeq Nano) used to produce reads of the newly sequenced taxa. The tribal/subtribal classification follows Soreng et al. (2014).

Tribe/subtribe Species Voucher specimens and GenBank accession numbers [ ] for newly sequenced plastomes, or place of plastome publication for previously sequenced plastomes Illumina library preparation method 
Brachyelytreae Brachyelytrum aristosum (Michx.) P. Beauv. ex Branner & Coville USA New York: J.I. Davis 777 (BH) [KM974735] TruSeq Nano 
Phaenospermateae Phaenosperma globosum Munro ex Benth. USA: J.I. Davis 779 (BH) [KM974745] TruSeq Nano 
Ampelodesmeae Ampelodesmos mauritanicus (Poir.) T. Durand & Schinz Germany. Royl & Schiers s.n. (B) [KM974731] Nextera 
Stipeae Achnatherum hymenoides (Roem. & Schult.) Barkworth Canada. British Columbia: J.M. Saarela, C.J. Sears & J.R. Maze 725 (CAN-590407) [KM974729] TruSeq 
 Oryzopsis asperifolia Michx. Canada. British Columbia: J.M. Saarela & D.M. Percy 430 (CAN-590301) [KM974744] TruSeq 
 Piptochaetium avenaceum (L.) Parodi USA Maryland: R.J. Soreng & K. Romaschenko (US) [KM974748] TruSeq Nano 
Meliceae Melica mutica Walter USA Maryland: W.J. Kress & M. Butts 04-7461 (US) [KM974742] TruSeq Nano 
 Melica subulata (Griseb.) Scribn. Canada. British Columbia: J.M. Saarela, D.M. Percy & Y.M. Chang 836 (CAN 590495) [KM974743] Nextera 
Diarrheneae Diarrhena obovata (Gleason) Brandenburg USA: J.I. Davis 756 (BH) [KM974739] TruSeq Nano 
Brachypodieae Brachypodium distachyon (L.) P. Beauv. Bortiri et al. (2008)  
Bromeae Bromus vulgaris (Hook.) Shear Canada. British Columbia: J.M. Saarela, D.M. Percy & Y.M. Chang 822 (CAN-590469) [KM974737] TruSeq 
Triticeae Aegilops cylindrica Host Middleton et al. (2014)  
 Aegilops geniculata Roth Middleton et al. (2014)  
 Aegilops speltoides Tausch Middleton et al. (2014)  
 Aegilops tauschii Coss. Middleton et al. (2014)  
 Hordeum jubatum L. Canada. Yukon: P.M. Peterson, J.M. Saarela & S.F. Smith 18478 (CAN-591095) [KM974741] TruSeq 
 Hordeum vulgare L. subsp. vulgare Saski et al. (2007)  
 Hordeum vulgare subsp. spontaneum (K. Koch) Asch. & Graebn. Middleton et al. (2014)  
 Secale cereale L. Middleton et al. (2014)  
 Triticum aestivum L. Bahieldin et al. (2014)  
 Triticum boeoticum Boiss. Middleton et al. (2014)  
 Triticum monococcum L. Middleton et al. (2014)  
 Triticum urartu Thumanjan ex Gandilyan Middleton et al. (2014)  
Poeae    
 Agrostidinae Agrostis stolonifera L. Saski et al. (2007)  
 Ammophila breviligulata Fernald USA New York: P.M. Peterson & J.M. Saarela 20867 (CAN) [KM974730] Nextera 
 Airinae Helictochloa hookeri (Scribn.) Romero Zarco Canada. Saskatchewan: P.M. Peterson, J.M. Saarela & S.F. Smith 18359 (CAN-590913) [KM974734] Nextera 
 Anthoxanthinae Anthoxanthum odoratum L. Canada. British Columbia: J.M. Saarela 500 (CAN-591412) [KM974732] TruSeq 
 Hierochloe odorata (L.) P. Beauv. USA Massachusetts: E.A Kellogg s.n. (A) [KM974740] Nextera 
 Aveninae Avena sativa L. Canada. British Columbia: J.M. Saarela & D.M. Percy 775 (CAN-590451) [KM974733] Nextera 
 Trisetum cernuum Trin. subsp. cernuum Canada: British Columbia: J.M. Saarela, D.M. Percy & Y.M. Chang 876 (CAN-0591575) [KM974753] Nextera 
 Brizinae s.s. Briza maxima L. Canada. Alberta [cultivated]: J.M. Saarela 284 (CAN) [KM974736] Nextera 
 Coleanthinae Puccinellia nuttalliana (Schult.) A.S. Hitchc. Canada. British Columbia: J.M. Saarela, C.J. Sears & J.R. Maze 713 (CAN 591508) [KM974750] TruSeq 
 Dactylidinae Dactylis glomerata L. Canada. British Columbia. J.M. Saarela 496 (CAN 591411) [KM974738] TruSeq 
 Holcinae Deschampsia antarctica E. Desv. Lee et al. (2014)  
 Loliinae Festuca altissima All. Hand et al. (2013)  
 Festuca ovina L. Hand et al. (2013)  
 Lolium multiflorum Lam. Hand et al. (2013)  
 Lolium perenne L. Diekmann et al. (2009)  
 Schedonorus arundinaceus (Schreb.) Dumort. Canada. British Columbia: J.M. Saarela 331 (CAN-591322) [KM974751] TruSeq 
 Schedonorus arundinaceus (Schreb.) Dumort. (as Lolium arundinaceum (Schreb.) Darbysh.) Cahoon et al. (2010)  
 Schedonorus pratensis (Huds.) P. Beauv. (as Festuca pratensis Huds.) Hand et al. (2013)  
 Phalaridinae Phalaris arundinacea L. Canada. British Columbia: J.M. Saarela, D.M. Percy & Y.M. Chang 973 (CAN-590547) [KM974746] TruSeq 
 Poinae Phleum alpinum L. Canada. British Columbia: J.M. Saarela & D.M. Percy 1234 (CAN-590642) [KM974747] TruSeq 
 Poa palustris L. Canada. British Columbia: J.M. Saarela & D.M. Percy 1080 (CAN-591589) [KM974749] TruSeq 
 Torreyochloinae Torreyochloa pallida var. pauciflora (J. Presl) J.I. Davis Canada. British Columbia: J.M. Saarela & D.M. Percy 1110 (CAN-591622) [KM974752] TruSeq 
Tribe/subtribe Species Voucher specimens and GenBank accession numbers [ ] for newly sequenced plastomes, or place of plastome publication for previously sequenced plastomes Illumina library preparation method 
Brachyelytreae Brachyelytrum aristosum (Michx.) P. Beauv. ex Branner & Coville USA New York: J.I. Davis 777 (BH) [KM974735] TruSeq Nano 
Phaenospermateae Phaenosperma globosum Munro ex Benth. USA: J.I. Davis 779 (BH) [KM974745] TruSeq Nano 
Ampelodesmeae Ampelodesmos mauritanicus (Poir.) T. Durand & Schinz Germany. Royl & Schiers s.n. (B) [KM974731] Nextera 
Stipeae Achnatherum hymenoides (Roem. & Schult.) Barkworth Canada. British Columbia: J.M. Saarela, C.J. Sears & J.R. Maze 725 (CAN-590407) [KM974729] TruSeq 
 Oryzopsis asperifolia Michx. Canada. British Columbia: J.M. Saarela & D.M. Percy 430 (CAN-590301) [KM974744] TruSeq 
 Piptochaetium avenaceum (L.) Parodi USA Maryland: R.J. Soreng & K. Romaschenko (US) [KM974748] TruSeq Nano 
Meliceae Melica mutica Walter USA Maryland: W.J. Kress & M. Butts 04-7461 (US) [KM974742] TruSeq Nano 
 Melica subulata (Griseb.) Scribn. Canada. British Columbia: J.M. Saarela, D.M. Percy & Y.M. Chang 836 (CAN 590495) [KM974743] Nextera 
Diarrheneae Diarrhena obovata (Gleason) Brandenburg USA: J.I. Davis 756 (BH) [KM974739] TruSeq Nano 
Brachypodieae Brachypodium distachyon (L.) P. Beauv. Bortiri et al. (2008)  
Bromeae Bromus vulgaris (Hook.) Shear Canada. British Columbia: J.M. Saarela, D.M. Percy & Y.M. Chang 822 (CAN-590469) [KM974737] TruSeq 
Triticeae Aegilops cylindrica Host Middleton et al. (2014)  
 Aegilops geniculata Roth Middleton et al. (2014)  
 Aegilops speltoides Tausch Middleton et al. (2014)  
 Aegilops tauschii Coss. Middleton et al. (2014)  
 Hordeum jubatum L. Canada. Yukon: P.M. Peterson, J.M. Saarela & S.F. Smith 18478 (CAN-591095) [KM974741] TruSeq 
 Hordeum vulgare L. subsp. vulgare Saski et al. (2007)  
 Hordeum vulgare subsp. spontaneum (K. Koch) Asch. & Graebn. Middleton et al. (2014)  
 Secale cereale L. Middleton et al. (2014)  
 Triticum aestivum L. Bahieldin et al. (2014)  
 Triticum boeoticum Boiss. Middleton et al. (2014)  
 Triticum monococcum L. Middleton et al. (2014)  
 Triticum urartu Thumanjan ex Gandilyan Middleton et al. (2014)  
Poeae    
 Agrostidinae Agrostis stolonifera L. Saski et al. (2007)  
 Ammophila breviligulata Fernald USA New York: P.M. Peterson & J.M. Saarela 20867 (CAN) [KM974730] Nextera 
 Airinae Helictochloa hookeri (Scribn.) Romero Zarco Canada. Saskatchewan: P.M. Peterson, J.M. Saarela & S.F. Smith 18359 (CAN-590913) [KM974734] Nextera 
 Anthoxanthinae Anthoxanthum odoratum L. Canada. British Columbia: J.M. Saarela 500 (CAN-591412) [KM974732] TruSeq 
 Hierochloe odorata (L.) P. Beauv. USA Massachusetts: E.A Kellogg s.n. (A) [KM974740] Nextera 
 Aveninae Avena sativa L. Canada. British Columbia: J.M. Saarela & D.M. Percy 775 (CAN-590451) [KM974733] Nextera 
 Trisetum cernuum Trin. subsp. cernuum Canada: British Columbia: J.M. Saarela, D.M. Percy & Y.M. Chang 876 (CAN-0591575) [KM974753] Nextera 
 Brizinae s.s. Briza maxima L. Canada. Alberta [cultivated]: J.M. Saarela 284 (CAN) [KM974736] Nextera 
 Coleanthinae Puccinellia nuttalliana (Schult.) A.S. Hitchc. Canada. British Columbia: J.M. Saarela, C.J. Sears & J.R. Maze 713 (CAN 591508) [KM974750] TruSeq 
 Dactylidinae Dactylis glomerata L. Canada. British Columbia. J.M. Saarela 496 (CAN 591411) [KM974738] TruSeq 
 Holcinae Deschampsia antarctica E. Desv. Lee et al. (2014)  
 Loliinae Festuca altissima All. Hand et al. (2013)  
 Festuca ovina L. Hand et al. (2013)  
 Lolium multiflorum Lam. Hand et al. (2013)  
 Lolium perenne L. Diekmann et al. (2009)  
 Schedonorus arundinaceus (Schreb.) Dumort. Canada. British Columbia: J.M. Saarela 331 (CAN-591322) [KM974751] TruSeq 
 Schedonorus arundinaceus (Schreb.) Dumort. (as Lolium arundinaceum (Schreb.) Darbysh.) Cahoon et al. (2010)  
 Schedonorus pratensis (Huds.) P. Beauv. (as Festuca pratensis Huds.) Hand et al. (2013)  
 Phalaridinae Phalaris arundinacea L. Canada. British Columbia: J.M. Saarela, D.M. Percy & Y.M. Chang 973 (CAN-590547) [KM974746] TruSeq 
 Poinae Phleum alpinum L. Canada. British Columbia: J.M. Saarela & D.M. Percy 1234 (CAN-590642) [KM974747] TruSeq 
 Poa palustris L. Canada. British Columbia: J.M. Saarela & D.M. Percy 1080 (CAN-591589) [KM974749] TruSeq 
 Torreyochloinae Torreyochloa pallida var. pauciflora (J. Presl) J.I. Davis Canada. British Columbia: J.M. Saarela & D.M. Percy 1110 (CAN-591622) [KM974752] TruSeq 

Plastome sequencing

Two micrograms of total genomic DNA from Brachyelytrum aristosum, Phaenosperma globosum, Piptochaetium avenaceum, Melica mutica and Diarrhena obovata were used in each library preparation. Libraries were prepared using the TruSeq Nano DNA sample preparation kit (Illumina, San Diego, CA, USA) and sequenced paired-end at Cold Spring Harbor Laboratory, Cold Spring, NY, USA. Libraries were prepared for the remaining taxa using the TruSeq and Nextera library preparation kits (Table 2). Detailed protocols for TruSeq, Nextera and TruSeq Nano are provided in Wysocki et al. (2014), Burke et al. (2014) and Barrett et al. (2013, 2014), respectively.

All Illumina-sequenced reads were first quality filtered using DynamicTrim v2.1 from the SolexaQA software package (Cox et al. 2010) with default settings, and then sequences <25 bp in length (default setting) were removed with LengthSort v2.1 in the same package.

Plastome assembly, annotation and alignment

Plastome assembly was performed entirely de novo. The Velvet software package (Zerbino and Birney 2008) was run iteratively following the methods from Wysocki et al. (2014). Contigs were scaffolded using the anchored conserved region extension method (Wysocki et al. 2014). Any remaining gaps in the plastomes were repaired by locating overlapping regions of 20 bp or higher using contigs or raw reads until the circular map was complete. Fully assembled plastomes were annotated by aligning to a previously published and annotated reference plastome in Geneious Pro (Biomatters Ltd, Auckland, New Zealand) and copying the reference annotations to the assembled plastome when the annotation shared a minimum similarity of 70 %. The banked plastome from Lolium perenne (NC009950) was used as an annotation reference.

Plastomes were arranged with the large single-copy (LSC) region followed by the inverted repeat region B (IRb) and ending with the short single-copy (SSC) region. Inverted repeat region A (IRa) was omitted from the matrix to prevent overrepresentation of the inverted repeat sequence. All newly assembled plastomes were then aligned, along with 20 previously published pooid plastomes and one bamboo outgroup plastome, Bambusa bambos (KJ870988) (Table 2), using the MAFFT alignment software (Katoh et al. 2005). We used the corrected sequence of Triticum aestivum (Bahieldin et al. 2014), as the earlier plastome (Ogihara et al. 2000) contained sequences from the rice plastome. The alignment was then inspected for structural mutations and adjusted manually to preserve tandem repeat boundaries and to identify inversions. Regions that contained inversion mutations were deleted from the matrix to remove false homology inferences. The alignment file may be obtained from TreeBase (http://purl.org/phylo/treebase/phylows/study/TB2:S16741).

Indels

Indel mutations were scored in the plastome alignment as in Leseberg and Duvall (2009). We scored indels that (i) were ≥2 bp long, excluding mono-nucleotide repeats regardless of length; (ii) could be attributed to slipped-strand mispairing, identified by the presence of a perfect or near-perfect repeated sequence; (iii) were unambiguous and did not overlap with other indels in the alignment; and (iv) were present in two or more individuals (i.e. autapomorphic indels were not scored). These criteria allowed us to focus on indels that could be interpreted as single evolutionary events. Indels were not included in our analyses. Based on the topology of the ML tree, scored indels were interpreted as representing putative synapomorphies or putative homoplasy.

Phylogeny estimation

We conducted maximum parsimony (MP), ML and Bayesian inference (BI) analyses with two different taxon sets and two subsets of the data. The first taxon set included 45 of the 46 taxa in our matrix, excluding a previously published plastome of Schedonorus arundinaceus that has been reported as having some sequencing errors (Hand et al. 2013), and the second included this plastome (46 taxa). For each of these matrices, we conducted analyses of complete plastomes (non-coding and protein coding) and protein-coding sequence (cds). The procedures, analyses and parameters noted below were used for all analyses. To exclude the potential alignment ambiguity from the analyses, nucleotide positions that contained one or more gaps introduced by the alignment were omitted from the matrix. The Akaike Information Criterion (AIC) was used in jModelTest v 2.1.3 (Guindon and Gascuel 2003; Darriba et al. 2012) to compare models of character evolution in each of the gap-free matrices. The General Time Reversible (GTR) model of substitution incorporating invariant sites and a γ distribution (GTR + I + G) was among the best-fit models and was used in subsequent analyses. Maximum likelihood analysis was performed using RAxML v 8.0.5 (Stamatakis 2006) with 1000 non-parametric bootstrap replicates. Non-parametric bootstrap values were generated using the Consense function in Phylip (Felsenstein 2005). MrBayes 3.2.2 (Ronquist and Huelsenbeck 2003) was used to perform the BI analyses. A Dirichlet prior was used for base frequencies and the rate matrix, and a uniform prior was used for the shape parameter (α), proportion of invariable sites (I) and topology; these are the default prior settings. Branch lengths were unconstrained and a GTR + I + G model was used with four discrete rate categories. The Metropolis-coupled Markov chain Monte Carlo (MCMCMC) search/sampling analysis was run for 2 × 10 000 000 generations with four chains. Average standard deviation of split frequencies remained <0.001 after the 50 % burn-in. Heuristic parsimony [1000 random addition sequence (RAS) replicates; TBR branch swapping; best trees kept] and parsimony bootstrap analysis with 1000 pseudoreplicates, 10 RAS each, were performed with PAUP* v4.0b10 (Swofford 2003). All results are presented on ML phylograms.

Testing outgroup selection effects

The set of 45 pooid plastomes was analysed with a randomized set of outgroup taxa to test the effect of outgroup selection on the tree topology. Outgroup taxa were selected from a pool of previously published plastome sequences from Bambusoideae [Acidosasa purpurea (NC015820), Arundinaria gigantea (NC020341), A. fargesii (NC024712), Bambusa bambos (KJ870988), Dendrocalamus latiflorus (NC013088), Fargesia nitida (NC024715), Ferrocalamus rimosivaginus (NC015831), Indocalamus longiauritus (NC015803), Phyllostachys nigra (NC015826), Olyra latifolia (KF515509)], Ehrhartoideae [Leersia tisserantii (NC016677), Oryza sativa (NC001320), Rhynchoryza subulata (NC016718)] and the PACMAD clade [Coix lacryma-jobi (NC013273), Neyraudia reynaudiana (NC024262), Panicum virgatum (NC015990), Sorghum bicolor (NC008602), Saccharum hybrid (NC006084) and Zea mays (NC001666)]. A representative species from each of the three lineages was chosen randomly using a custom Python script (available on request), and aligned with the pooid plastomes with the MAFFT method. Phylogeny was estimated in an ML framework using RAxML. This was repeated 16 times and the tree with the highest likelihood from each iteration was tested for congruence using the Consense function of the Phylip software package.

Results

Plastome sequencing

Complete plastomes were newly sequenced for 25 pooid grass species. Plastome lengths ranged from 134 287 to 137 897 bp. All lengths of plastomes and their sub-regions are reported in Table 3. After the removal of one inverted repeat region, the 46-taxon alignment included 135 838 nucleotide positions, which decreased to 94 022 positions after the removal of all positions with at least one introduced gap (30.78 % positions removed). The 45-taxon alignment included 135 059 nucleotide positions, which decreased to 94 209 positions after the removal of all positions with at least one introduced gap (30.25 % positions removed). Removing all gapped positions reduced the 46-taxon protein-coding region alignment from 47 001 to 45 213 positions (3.80 % removed) and reduced the 45-taxon alignment from 46 708 to 45 256 positions (3.11 % removed).

Table 3.

Lengths (bp) of newly sequenced plastomes and their sub-regions. LSC, large single-copy; IR, inverted repeat; SSC, short single-copy.

Taxon Total LSC IR SSC 
Achnatherum hymenoides 137 742 81 709 21 615 12 803 
Ammophila breviligulata 136 725 80 710 21 657 12 701 
Ampelodesmos mauritanicus 136 975 79 543 22 273 12 886 
Anthoxanthum odoratum 135 551 79 626 21 627 12 671 
Avena sativa 135 890 80 109 21 603 12 575 
Brachyelytrum aristosum 137 399 81 819 21 434 12 712 
Briza maxima 136 823 79 707 22 197 12 722 
Bromus vulgaris 136 935 80 964 21 702 12 567 
Dactylis glomerata 134 737 79 523 21 472 12 270 
Diarrhena obovata 137 421 81 367 21 621 12 812 
Hierochloe odorata 136 394 80 645 21 642 12 465 
Helictochloa hookeri 134 976 79 370 21 502 12 602 
Hordeum jubatum 136 834 80 901 21 629 12 675 
Melica mutica 134 710 80 478 20 831 12 570 
Melica subulata 134 773 80 411 20 836 12 690 
Oryzopsis asperifolia 134 287 80 475 20 503 12 806 
Phaenosperma globosum 137 897 82 128 21 446 12 877 
Phalaris arundinacea 135 873 79 833 21 530 12 980 
Phleum alpinum 135 568 80 009 21 368 12 823 
Piptochaetium avenaceum 137 701 81 613 21 625 12 838 
Poa palustris 135 446 79 566 21 552 12 776 
Puccinellia nuttalliana 135 352 79 594 21 516 12 726 
Schedonorus arundinaceus 135 266 79 934 21 421 12 490 
Torreyochloa pallida 136 102 80 137 21 641 12 683 
Trisetum cernuum 135 539 79 828 21 636 12 439 
Taxon Total LSC IR SSC 
Achnatherum hymenoides 137 742 81 709 21 615 12 803 
Ammophila breviligulata 136 725 80 710 21 657 12 701 
Ampelodesmos mauritanicus 136 975 79 543 22 273 12 886 
Anthoxanthum odoratum 135 551 79 626 21 627 12 671 
Avena sativa 135 890 80 109 21 603 12 575 
Brachyelytrum aristosum 137 399 81 819 21 434 12 712 
Briza maxima 136 823 79 707 22 197 12 722 
Bromus vulgaris 136 935 80 964 21 702 12 567 
Dactylis glomerata 134 737 79 523 21 472 12 270 
Diarrhena obovata 137 421 81 367 21 621 12 812 
Hierochloe odorata 136 394 80 645 21 642 12 465 
Helictochloa hookeri 134 976 79 370 21 502 12 602 
Hordeum jubatum 136 834 80 901 21 629 12 675 
Melica mutica 134 710 80 478 20 831 12 570 
Melica subulata 134 773 80 411 20 836 12 690 
Oryzopsis asperifolia 134 287 80 475 20 503 12 806 
Phaenosperma globosum 137 897 82 128 21 446 12 877 
Phalaris arundinacea 135 873 79 833 21 530 12 980 
Phleum alpinum 135 568 80 009 21 368 12 823 
Piptochaetium avenaceum 137 701 81 613 21 625 12 838 
Poa palustris 135 446 79 566 21 552 12 776 
Puccinellia nuttalliana 135 352 79 594 21 516 12 726 
Schedonorus arundinaceus 135 266 79 934 21 421 12 490 
Torreyochloa pallida 136 102 80 137 21 641 12 683 
Trisetum cernuum 135 539 79 828 21 636 12 439 

Testing outgroup selection effects

Sixteen iterations of randomly sampling the outgroup taxa produced 16 identical ingroup topologies. However, the maximum likelihood bootstrap support (MLBS) did vary at five nodes. The sister group to Melica showed a mean MLBS of 99.75 (±0.58), the sister group to Diarrhena obovata showed a mean MLBS of 82.19 (±5.94), the sister group to the Phalaris–Torreyochloa clade showed a mean bootstrap support (BS) of 93.38 (±2.16), the sister group to Helictochloa hookeri showed a mean BS of 74.38 (±13.09) and the sister group to Deschampsia antarctica showed a mean BS of 80.63 (±12.83).

Phylogeny

The 45-taxon MP analyses of the complete and protein cds resulted in one and two equally most parsimonious trees, respectively. In the 45-taxon analysis of the complete data, all but four nodes in the ML tree and three nodes in the MP tree are maximally supported (BS = 100 %) and all nodes in the BI tree have posterior probabilities of 1 (Fig. 1). In the 45-taxon analysis of protein cds, all but four nodes in the ML tree and three nodes in the BI tree were maximally supported (Fig. 2). In the MP tree, five nodes received support between 50 and 99 %, and no topology among Helictochloa, Dactylis, Deschampsia and Loliinae received >50 % BS (Fig. 2). Support is lower in the protein cds analyses compared with the complete analyses at nodes that are not maximally supported in both. The topologies of the ML and BI trees are identical in the 45-taxon analyses of both data subsets. The parsimony trees differ in the relative placements of four subtribes (two clades) in one major clade of tribe Poeae, and two to three subtribes in a second major clade of the tribe. Bootstrap support in parsimony trees is generally lower than BS in ML trees at nodes that are not maximally supported in both.

Figure 1.

Maximum likelihood phylogram of complete plastomes of 44 pooid grasses and one outgroup taxon. Tribes of Pooideae, subtribes of Poeae and Poeae clades 1 and 2 are indicated. Bootstrap values are indicated only when at least one is less than maximally supported (ML bootstrap value precedes MP bootstrap value). ML and Bayesian (BI) topologies are identical. Posterior probabilities at all nodes = 1.00. ‘NR’ indicates a node not resolved or supported above the 50 % bootstrap level in the MP analysis. Crosses indicate clades that are in reversed positions in the MP tree. The MP bootstrap value of the node affected by the reversal of these clades reflects the value associated with the reversed topology, i.e. the topology not reflected in this figure.

Figure 1.

Maximum likelihood phylogram of complete plastomes of 44 pooid grasses and one outgroup taxon. Tribes of Pooideae, subtribes of Poeae and Poeae clades 1 and 2 are indicated. Bootstrap values are indicated only when at least one is less than maximally supported (ML bootstrap value precedes MP bootstrap value). ML and Bayesian (BI) topologies are identical. Posterior probabilities at all nodes = 1.00. ‘NR’ indicates a node not resolved or supported above the 50 % bootstrap level in the MP analysis. Crosses indicate clades that are in reversed positions in the MP tree. The MP bootstrap value of the node affected by the reversal of these clades reflects the value associated with the reversed topology, i.e. the topology not reflected in this figure.

Figure 2.

Maximum likelihood phylogram of plastome protein cds of 44 pooid grasses and 1 outgroup taxon. Tribes of Pooideae, subtribes of Poeae and Poeae clades 1 and 2 are indicated. Support values are indicated only when at least one is less than maximally supported (ML bootstrap value precedes MP bootstrap value, which precedes the Bayesian posterior probability). ‘NR’ indicates nodes not resolved or supported above the 50 % bootstrap level in one of the analyses. Crosses indicate clades that were reversed in the MP tree. The MP bootstrap value of the node affected by the reversal of clades reflects that of the reversed topology. The histogram indicates total invariant (stippled) and variable (solid) nucleotide sites in the full plastome analysis (Fig. 1) and this analysis.

Figure 2.

Maximum likelihood phylogram of plastome protein cds of 44 pooid grasses and 1 outgroup taxon. Tribes of Pooideae, subtribes of Poeae and Poeae clades 1 and 2 are indicated. Support values are indicated only when at least one is less than maximally supported (ML bootstrap value precedes MP bootstrap value, which precedes the Bayesian posterior probability). ‘NR’ indicates nodes not resolved or supported above the 50 % bootstrap level in one of the analyses. Crosses indicate clades that were reversed in the MP tree. The MP bootstrap value of the node affected by the reversal of clades reflects that of the reversed topology. The histogram indicates total invariant (stippled) and variable (solid) nucleotide sites in the full plastome analysis (Fig. 1) and this analysis.

In the 46-taxon analysis of the complete data, all but three nodes in the ML tree and four nodes in the MP tree are maximally supported, and all nodes in the BI tree are maximally supported [see Supporting Information]. In the 46-taxon analysis of protein cds, all but six nodes in the ML tree, seven nodes in the MP tree and four nodes in the BI tree are maximally supported. Support for most nodes is lower in the trees based on protein cds tree compared with the complete plastome data. The remainder of the text focusses on the 45-taxon dataset, unless indicated otherwise.

Ampelodesmeae is nested in Stipeae, and henceforth we refer to this lineage as the Ampelodesmeae–Stipeae clade. Within this clade, Achnatherum and Piptochaetium are successive sisters to an OryzopsisAmpelodesmos clade.

In all analyses, Brachyelytreae are inferred to be the sister group of the rest of Pooideae. Phaenospermateae, Meliceae, Ampelodesmeae–Stipeae and Diarrheneae are the next successively diverging lineages, sister to the core Pooideae. The core Pooideae including Brachypodieae, Bromeae, Poeae and Triticeae are a moderately to strongly supported clade in all analyses of the complete data [MLBS = 89 %, maximum parsimony bootstrap support (MPBS) = 100 %, PP = 1.00] (Fig. 1). Support is lower based on the protein cds (MLBS <50 %, MPBS = 81 %, PP = 0.86); in the ML tree, Diarrhena and Brachypodium are a weakly supported clade (MLBS = 56 %, data not shown) (Fig. 2).

Within the core Pooideae, Bromeae and Triticeae are sister taxa, and this clade is sister to Poeae. Within Triticeae, Secale is the sister group of an AegilopsTriticum clade in which neither Aegilops nor Triticum is monophyletic, and Hordeum is the sister group to this larger lineage. Poeae is divided into two clades: Poeae clade 1, including Agrostidinae, Brizinae s.s., Anthoxanthinae, Aveninae, Phalaridinae and Torreyochloinae; and Poeae clade 2, including Coleanthinae, Poinae, Airinae, Holcinae, Dactylidinae and Loliinae. In Poeae clade 1, one subclade has the following topology: [Anthoxanthinae, (Brizinae, Agrostidinae)] and Phalaridinae and Torreyochloinae form a clade. In the ML and BI trees, Phalaridinae–Torreyochloinae and Aveninae are strongly (MLBS = 94 %, PP = 1.00; Fig. 1) to weakly (MLBS = 59 %, PP = 0.72; Fig. 2) supported as the successive sister groups of the rest of the clade, whereas in the MP tree the branching order of these two lineages is reversed and weakly (MPBS = 69 %; Fig. 1) to strongly (MPBS = 91 %; Fig. 2) supported. In Poeae clade 2, Coleanthinae and Poinae are sister taxa, and a second subclade comprises Airinae, Holcinae, Dactylidinae and Loliinae. In ML and BI complete data trees Airinae and Holcinae are strongly supported successive sister groups of the rest of the subclade, whereas in the MP tree Airinae and Holcinae are an unsupported clade (Fig. 1); and relationships at the base of this clade are unresolved in the MP analysis of protein cds (Fig. 2). In ML and BI analyses, Dactylidinae and Loliinae are sister taxa; this relationship is strongly supported in analyses of the complete data but poorly supported in the protein-coding analyses.

Unique plastome features

The plastome from Brachyelytrum aristosum shares a 196 bp insertion with Bambusa bambos in the rps16-trnQ intergenic spacer. The two insertions exhibit 85 % identity. This insertion is not present in any other genera of Pooideae but is represented in all subfamilies of grasses except for Danthonioideae and Micrairoideae. Phaenosperma globosum and both species of Melica share a 198 bp insertion in the same region that shows no significant sequence similarity to the insertion shared by Bambusa bambos and Brachyelytrum aristosum. The insertions found in Melica and P. globosum are not found in any other species of Pooideae included here, but a BLAST query reveals similar insertions in one species of Puelioideae (Puelia olyriformis), an ‘early-diverging’ member of the Panicoideae (Thysanolaena latifolia) and many members of the temperate bamboo lineage Arundinarieae.

Two undocumented putative insertions of non-plastid homology were found in two previously published plastomes (Fig. 3). The full plastome for Triticum urartu contains a 1060 bp insertion in the region of the inverted repeat that ordinarily contains the coding sequence for the rpl23 gene. A BLAST query of this sequence against GenBank shows the region to have homology to various nuclear genes within multiple species of Triticum. Triticum monococcum contains a 1077 bp insertion in the rpl23-ndhB intergenic spacer of the inverted repeat region. A BLAST query of this sequence against GenBank shows the region to have homology to the Triticum timopheevii mitochondrial genome and many other grass mitochondrial regions. The top hit was located in the rrn26-1 region of the T. timopheevii mitochondrial genome, covered nearly 100 % of plastome insertion and exhibited over 99 % identity.

Figure 3.

Two putative insertions of nuclear and mitochondrial homology located in IR regions of the Triticum urartu (NC_021762) and T. monococcum (NC_021762) plastomes, respectively, are shown here and compared with the typical Hordeum jubatum plastome (KM974741). The insertion in the T. urartu plastome exhibits 90 % identity to chromosome 3B of the T. aestivum nuclear genome (HG670306) and the insertion in the T. monococcum plastome exhibits 99 % identity to the T. timopheevii mitochondrial genome (AP013106). Gene orientation is indicated and thin lines denote regions in which gaps were introduced to preserve the alignment. Gene position is relative and lengths are not to scale.

Figure 3.

Two putative insertions of nuclear and mitochondrial homology located in IR regions of the Triticum urartu (NC_021762) and T. monococcum (NC_021762) plastomes, respectively, are shown here and compared with the typical Hordeum jubatum plastome (KM974741). The insertion in the T. urartu plastome exhibits 90 % identity to chromosome 3B of the T. aestivum nuclear genome (HG670306) and the insertion in the T. monococcum plastome exhibits 99 % identity to the T. timopheevii mitochondrial genome (AP013106). Gene orientation is indicated and thin lines denote regions in which gaps were introduced to preserve the alignment. Gene position is relative and lengths are not to scale.

Indels

A total of 177 indels were scored that we interpreted as single evolutionary events [see Supporting Information]. Eighty-four of these could be attributed to slipped-strand mispairing. All of the scored indels could be straightforwardly interpreted as being putatively synapomorphic or putatively homoplastic. Given the current level of taxon sampling, 34 indels were putatively homoplastic with respect to the topology of the ML tree, while the remaining 143 represented putative synapomorphies for clades of various composition, ranging from multi-tribal clades to those of a single species (in cases where more than one individual per species was sampled) (Table 4). One indel was putatively synapomorphic for the large clade comprising Diarrheneae plus the core Pooideae; 2 indels each were putatively synapomorphic for a three-tribe and a four-tribe lineage; 30 indels were putatively synapomorphic for 6 two-tribe lineages, with 20 of these supporting the Bromeae–Triticeae clade; 37 indels were putatively synapomorphic for 8 tribes, with 13 of these supporting Meliceae and 7 supporting Triticeae and 9 indels were putatively synapomorphic for three species. Within Triticeae, 7 indels represented putative synapomorphies for clades of subsets of Aegilops and Triticum species, 4 were putatively synapomorphic for an AegilopsTriticum clade, 10 were putatively synapomorphic for an AegilopsTriticumSecale clade and 6 were putatively synapomorphic for Hordeum. Within Loliinae 4 indels were putatively synapomorphies for a clade of all sampled taxa except Festuca ovina, and 14 were putative synapomorphies for a LoliumSchedonorus clade. Seven and ten indels were putative synapomorphies for Poeae clades 1 and 2, respectively. A total of 111 indels are located in intergenic spacer regions, 33 in introns and 14 in protein-coding regions (genes).

Table 4.

Unambiguous indels scored in the 46-taxon plastome alignment that are putatively synapomorphic for clades of two or more taxa, or single taxa represented by more than one individual. Putatively homoplasious indels are also scored. [See Supporting Information] for details of scored indels.

Clade Number of indels 
Aegilops cylindricaA. tauschii 
Aegilops speltoidesTriticum aestivum 
AegilopsTriticum 
AegilopsTriticum minus A. speltoides and T. aestivum 
AegilopsTriticumSecale 10 
Agrostidinae 
Agrostidinae–Brizinae 
Airinae–Dactylidinae–Loliinae 
Ampelodesmeae–Stipeae 
Anthoxanthinae 
Aveninae 
Bromeae–Poeae–Triticeae 
Bromeae–Triticeae 20 
Coleanthinae–Poeae 
Diarrheneae–Brachypodieae–Bromeae–Poeae–Triticeae 
Homoplasious indels 34 
Hordeum 
Hordeum vulgare 
Loliinae 
LoliumSchedonorus 14 
LoliumSchedonorusFestuca altissima 
Melica 13 
Phalaridinae–Torreyochloinae 
Poeae 
Poeae clade 1 
Poeae clade 2 10 
Poinae 
Schedonorus arundinaceus 
Stipeae 
Triticeae 
Triticum minus T. aestivum 
Triticum monococcum 
Clade Number of indels 
Aegilops cylindricaA. tauschii 
Aegilops speltoidesTriticum aestivum 
AegilopsTriticum 
AegilopsTriticum minus A. speltoides and T. aestivum 
AegilopsTriticumSecale 10 
Agrostidinae 
Agrostidinae–Brizinae 
Airinae–Dactylidinae–Loliinae 
Ampelodesmeae–Stipeae 
Anthoxanthinae 
Aveninae 
Bromeae–Poeae–Triticeae 
Bromeae–Triticeae 20 
Coleanthinae–Poeae 
Diarrheneae–Brachypodieae–Bromeae–Poeae–Triticeae 
Homoplasious indels 34 
Hordeum 
Hordeum vulgare 
Loliinae 
LoliumSchedonorus 14 
LoliumSchedonorusFestuca altissima 
Melica 13 
Phalaridinae–Torreyochloinae 
Poeae 
Poeae clade 1 
Poeae clade 2 10 
Poinae 
Schedonorus arundinaceus 
Stipeae 
Triticeae 
Triticum minus T. aestivum 
Triticum monococcum 

Discussion

Phylogenomic analyses of 45 whole plastomes, including 25 newly sequenced here, resulted in a highly resolved and strongly supported phylogeny of Pooideae, with the caveat that whole plastomes for three tribes (Nardeae, Lygeae, Brylkineae) and Littledalea are not yet sampled. The few topological differences between the MP, ML and BI trees are in tribe Poeae and at the base of the core Pooideae. Each of the discordant inferred clades is subtended by a very short branch. As most branches in the Poeae clade define genera and are fairly long, the topological differences observed may be a reflection of long-branch attraction (LBA) in the MP analysis due to the proximity of very short and very long branches in the tree. This is a common problem in phylogenetic inference and has been observed in other whole plastome phylogenomic studies (Soltis and Soltis 2004; Stefanović et al. 2004; Givnish et al. 2010; Barrett et al. 2014). Strategies to overcome LBA include using inference methods that are less prone to long branch effects and, therefore, may be more accurate (such as ML or BI inference, which we have done here); excluding third codon positions, which may be saturated or randomized; representing clades in analyses with only short-branched taxa by excluding taxa with long branches; adding taxa to break up large branches and adding data (Bergsten 2005). An alternative strategy to adding more data could involve data filtration. For example, quality measures of ‘tree-likeness’ for data partitions could be used to identify the ‘data core.’ Aside from LBA, factors that might lead to differences in topology and/or support among data partitions include model mis-specification, intrinsic biases among the data (e.g. large differences in GC content among clades; Ruhfel et al. 2014) and heterotachy.

For most tribes and subtribes we have sampled only one or two exemplar taxa, so removing taxa from the analyses would be a poor approach. In contrast, with rapidly decreasing costs for sequencing whole plastomes, there is ample opportunity to increase taxon sampling within and among tribes, particularly those that are species-rich, such as Poeae—the clade in which the incongruences among analyses were found and in which there are numerous unresolved phylogenetic questions. Given our current strategy of sampling whole plastomes, adding more data from this linkage group is impossible, by definition. Adding data from the mitochondrial or nuclear genomes would be beneficial for deciphering the true evolutionary history of taxa (i.e. the species tree), as genes from these genomes may have evolutionary histories that are different than those of maternally inherited plastid genomes, such as the nuclear gene history found by Triplett et al. (2014), but this would not help with the long-branch problem in the plastid data. Coalescent-based analyses of multiple unlinked nuclear genes could help accurately reconstruct species relationships, especially for parts of the plastome phylogeny that may be at odds with the genealogies of some or many nuclear genes. Because probability-based methods of phylogenetic inference are less prone to long-branch effects than parsimony, our discussion below is based primarily on the topologies of our nearly identical ML and BI trees, which also are generally better supported than the MP trees.

Previous phylogenomic studies of relationships among orders and families of plants have examined only protein-coding genes (e.g. Barrett et al. 2013, 2014; Davis et al. 2013; Kim and Kim 2013; Martin et al. 2013; Ruhfel et al. 2014), while those—including the current study—that have included non-coding data have generally focussed on more recent or ‘shallow’ phylogenetic relationships (Zhang et al. 2011; Burke et al. 2012; Hand et al. 2013; Gornicki et al. 2014; Huang et al. 2014; Ma et al. 2014; Middleton et al. 2014). Even though spacers/introns are ‘non-coding’ some plastid intergenic regions are expected to be conserved due to containing enzyme-binding sites or important secondary structures (Peredo et al. 2012) and they may too be under purifying selection similar to coding loci, and thus useful in larger-scale phylogenetic reconstructions. Nevertheless, aligning non-coding regions (or a subset of them) across divergent taxa can be difficult and may introduce error into phylogenetic analyses when alignments are not accurate, and we have attempted to minimize the possibility of this type of error by excluding from analysis all positions in the alignment that include a gap in one or more taxon. The improvement in support in our analyses of complete plastomes versus protein cds indicates that the non-coding complement of the genome contributes important characters for resolving these mostly deep nodes in pooid grasses, similar to the results of other phylogenomic studies of grasses (Burke et al. 2012; Ma et al. 2014) and other plants (Eserman et al. 2014) in which analyses of complete plastomes and/or non-coding data and protein cds were explicitly compared. Inclusion of non-coding data in plastome phylogenomic analyses of such groups as Zingiberales (Barrett et al. 2014) may improve support levels at deep nodes that are weakly supported based on protein cds alone.

‘Early-diverging’ Pooideae

Among the tribes of Pooideae major uncertainties of relationship have revolved around the relative branching orders of Phaenospermateae, Brylkinieae–Meliceae and Ampelodesmeae–Stipeae following the respective divergences of Brachyelytrum and Lygeae–Nardeae. Plastomes are not yet available for Lygeum and Nardus. As expected, Brachyelytrum represents one of the lineages resulting from the deepest split in the subfamily, and we find strong support for Phaenospermateae, Brylkinieae–Meliceae and Ampelodesmeae–Stipeae to be the next successive sister groups to the rest of Pooideae. The branch subtending all Pooideae except Brachyelytrum is quite long, indicating a long period of time prior to the next divergence and/or a rapid rate of plastid evolution along this branch. Conversely, the next two deep branches at the base of the plastome trees identifying Phaenospermateae, Meliceae and Ampelodesmeae–Stipeae as successively diverging taxa are extremely short compared with most other deep branches in the tree subtending lineages recognized as tribes, possibly reflecting rapid radiation of these lineages. Given these short branches, it is not surprising that most previous studies were not able to robustly resolve the relationships among these taxa, particularly those based on one or a few gene regions (Catalán et al. 1997; Mathews et al. 2000; Döring et al. 2007; Schneider et al. 2009; Davis and Soreng 2010; Blaner et al. 2014). Our results are congruent with the largest study, in terms of gene regions sampled, that sampled these lineages (Romaschenko et al. 2012), although not all nodes in that study were strongly supported as they are in our plastome trees. In contrast, our results are discordant with two- and three-gene studies that recovered strong support in BI analyses for the opposite order of divergence for Ampelodesmeae–Stipeae and Meliceae (Duvall et al. 2007; Bouchenak-Khelladi et al. 2008). The reasons for this are not clear, but may be due to differing levels of taxon sampling in and near this part of the tree, outgroup choice (although tests of different grass outgroups here had no effect on the ingroup topology) or varying phylogenetic signal in different partitions of the data [as demonstrated for monocots in Davis et al. (2013), for example]. Considering the latter possibility, support was much lower for the current topology in the MP analyses of protein-coding data compared with the complete dataset, while support in the ML and BI analyses was strong in both. Diarrheneae represent the next successive sister group to the rest of the subfamily (the core Pooideae) in most of our plastome trees, congruent with the results of other studies [Clark et al. 1995; Catalán et al. 1997; Davis and Soreng 2007, 2010; Döring et al. 2007; Duvall et al. 2007 (Bayesian analysis)]. An alternative topology recovered only in the ML analysis of protein cds, in which Brachypodium and Diarrhena are a clade, was poorly supported. This topology was also found in Minaya et al. (2013), with strong support.

Core Pooideae

Brachypodieae

Accurate placement of Brachypodium in the context of the phylogeny of Pooideae is important, as B. distachyon has become a model species for the genomics of cereal crops, grass biofuels and forage grasses (The International Brachypodium Initiative 2010; Mochida and Shinozaki 2013). Relationships among the four sampled core tribes of Pooideae inferred by whole plastomes are consistent with most previous plastid trees, in which Brachypodieae is the sister group of a clade comprising a Bromeae–Triticeae lineage and Poeae (Catalán et al. 1997; Hilu et al. 1999; Duvall et al. 2007; Schneider et al. 2009; Davis and Soreng 2010; Grass Phylogeny Working Group II 2012; Blaner et al. 2014). This topology is also present in the nuclear topoisomerase 6 phylogeny of Blaner et al. (2014). Two other nuclear-based phylogenies, however, infer alternative highly supported topologies that are discordant with each other and with the plastome phylogeny. The phytochrome B tree of Mathews et al. (2000) identifies Diarrhena as the sister group to the core Pooideae (excluding Brachypodium), with Brachypodium sister to this clade, whereas the β-amylase tree of Minaya et al. (2013) identifies a strongly supported (Brachypodieae + Diarrheneae), [Stipeae, (Lygeae + Nardeae)] clade that is sister to the rest of the core Poeae. The combined nuclear (ITS, β-amylase) and plastid tree in Minaya et al. (2013), however, has the same topology as our plastome tree with respect to the relationships among the core Pooideae, with all the relevant nodes strongly supported. This level of support is surprising given the strongly discordant β-amylase tree, which was incongruent with the ITS tree in a partition homogeneity test, as were the nuclear and plastid data (Minaya et al. 2013). Unfortunately, the ITS and plastid trees were not reported in that study and we are therefore unable to compare them with the β-amylase tree.

Additional nuclear genes have been sequenced for Brachypodium and representatives of Triticeae and Meliceae in the context of understanding the evolutionary history of Brachypodium species, but these studies did not have sufficient sampling to address the affinities of Brachypodium with respect to other Pooideae (Wolny et al. 2011; Catalán et al. 2012). Similarly, gene trees of nuclear loci coding for economically important traits, such as grain endosperm texture (Charles et al. 2009) have been studied, but these are based on data from the few sequenced grass genomes (rice, sorghum, wheat, Brachypodium), a level of taxon sampling that is insufficient for reconstructing finer-scale aspects of phylogeny. Although there is consensus on the position of Brachypodium in plastid-based phylogenies as the sister group of the rest of the core Pooideae, additional nuclear-based phylogenies with dense sampling are needed to distinguish gene trees from the species tree to characterize the precise affinities of Brachypodium and other pooid grasses, given the discordances among existing nuclear gene trees.

In the current study, the single plastome representative of Brachypodieae, the annual species Brachypodium distachyon, lies on a very long branch and is subtended by a very short one. Although the topologies are the same in five of the six MP, ML and BI trees, the branch length may be distorting the analyses. Available plastid data indicate that B. distachyon has the highest substitution rate in the genus. For example, within the crown Brachypodium clade in an ndhF phylogeny B. distachyon lies on a branch two and a half to eight times the length of the terminal branches of other taxa in the genus, which are perennials (Catalán et al. 1997). Rate variation between annual and perennial plant taxa has been widely documented (e.g. Yue et al. 2010; Gaut et al. 2011) and was recently demonstrated in grasses in plastome phylogenies of woody vs. herbaceous bamboos (Burke et al. 2012, 2014). Until recently B. distachyon was considered to be the only annual species in its genus, but the taxon has now been split into three annual species on the basis of morphological, cytogenetic and molecular data (Catalán et al. 2012). The plastome included here was sequenced from the Bd21 (2n = 10) genotype of Brachypodium (Bortiri et al. 2008) and corresponds to B. distachyon s.s. in the revised classification. Although an earlier phylogenetic analysis identified B. distachyon (s.l.) as the sister group of the rest of the genus (Catalán and Olmstead 2000), a more detailed study identifies two perennial species and one of the newly described annuals (B. stacei) as representing the first (but poorly supported) splits in the genus, with B. distachyon s.s. and the other newly described annual, B. hybridum, sister to a clade of the remaining perennial taxa (Catalán et al. 2012). Sampling plastomes from one of the core perennial species and one or more of the putative ‘early-diverging’ taxa would contribute to an understanding of plastome variation in the lineage and might break up the long branch in the crown Brachypodium clade, which may improve support in analyses of protein cds, and may possibly affect support levels elsewhere in the tree.

Poeae clade 1

As in previous plastid studies, the tribe Poeae is divided into two strongly supported clades, Poeae clades 1 and 2, that are further supported by 7 and ten unambiguous indels, respectively. The plastome phylogenies of Poeae clade 1 are maximally supported at all nodes in the ML and BI trees and at all but one node in the MP trees, and the topologies of the ML/BI and MP trees differ only in the relative placements of Aveninae and Phalaridinae–Torreyochloinae as the sister group of the rest of the clade. Our sampling represents all six subtribes that are part of Poeae clade 1 in Soreng et al. (2014). Brizinae is divided into two informal groups in the classification: Brizinae s.s., which includes the Old World genera Airopsis and Briza s.s., and Brizinae s.l. ‘Calotheca clade’, which includes the New World genera Chascolytrum s.l. (see Essi et al. 2011 for details on the recent expansion of Chascolytrum) and Relchela. This division is based on the results of phylogenetic work that shows Brizinae to be non-monophyletic. In an early study, Soreng and Davis (2000) found Brizinae to be paraphyletic and suggested that the subtribe represents parallel evolution of a ‘brizoid’ lemma (i.e. lemmas that are as long as broad) in Eurasia and South America. This paraphyly was supported by numerous subsequent phylogenetic studies: species of Chascolytrum s.l. and subtribe Agrostidinae form a clade, and Briza s.s. and Airopsis represent a distinct lineage that, in most analyses, is the sister group of the former clade (Davis and Soreng 2007; Döring et al. 2007; Quintanar et al. 2007; Soreng et al. 2007; Saarela et al. 2010; Grass Phylogeny Working Group II 2012). Chascolytrum s.l. is anomalous morphologically in the Agrostidinae clade as it has multiple florets per spikelet, compared with a single floret per spikelet, which is the diagnostic character for the subtribe. The distinction between these Old World and New World Brizinae lineages is clearly shown in Essi et al. (2008), although that study unfortunately did not include any closely related non-Brizinae taxa and thus the two Brizinae lineages appear to be reciprocally monophyletic sister taxa, which may not be accurate. Although we have not sampled the Brizinae s.l. ‘Calotheca clade’ here, our plastome trees similarly identify Brizinae s.s. as the sister group of Agrostidinae, with improved support compared with most earlier studies. Characterizing relationships between the Brizinae s.l. ‘Calotheca clade’ (i.e. Chascolytrum s.l.) and Agrostidinae taxa will require further work. The former group has been resolved as monophyletic in some, but not all studies, and its affinities with Agrostidinae taxa have been variously resolved (Davis and Soreng 2007; Quintanar et al. 2007; Soreng et al. 2007; Saarela et al. 2010). The monotypic Relchela has only been sampled in one study, where in an ITS tree it falls in a clade of Agrostidinae taxa (Refulio-Rodriguez 2007). We have sampled only two (Agrostis, Ammophila) of the ∼16 genera recognized in Agrostidinae, and these form a strongly supported clade. Other studies have similarly identified Agrostidinae as a monophyletic group (often including Chascolytrum s.l., as noted above) (Soreng and Davis 2000; Davis and Soreng 2007; Döring et al. 2007; Quintanar et al. 2007; Minaya et al. 2013) but further research is needed to clarify generic concepts and relationships within the subtribe (Saarela et al. 2010).

The strongly supported sister group relationship in our plastome trees between Anthoxanthinae and Brizinae–Agrostidinae has been found previously only in two plastid trees, in which this topology was weakly supported (Bouchenak-Khelladi et al. 2008; Saarela et al. 2010). This topology is incongruent with the placement of Anthoxanthinae in a strongly supported clade with Aveninae in other plastid, nuclear and combined trees (Davis and Soreng 2007; Döring et al. 2007; Quintanar et al. 2007; Schneider et al. 2009; Minaya et al. 2013) or sister to Phalaris (Soreng and Davis 2000). The reasons for these varying topologies are unclear, but may be related to density of taxon sampling, variable signal in different data partitions and/or rates of evolution in Poeae clade 1.

The strongly supported sister group relationship between Phalaridinae and Torreyochloinae in our plastome trees has not been identified previously. This may be because both lineages have been sampled together in only a few analyses, at least when considering the current circumscription of Phalaridinae. This subtribe has traditionally been circumscribed as including two or three genera, the closely related Anthoxanthum and Hierochloe (often included in Anthoxanthum) and Phalaris, which have a similar and unique spikelet structure in Poeae (Clayton and Renvoize 1986; Soreng et al. 2007). However, this circumscription has not been supported by most molecular studies—including the current one—which have identified Phalaris and Anthoxanthum s.l. as separate lineages (Döring et al. 2007; Quintanar et al. 2007; Bouchenak-Khelladi et al. 2008; Gillespie et al. 2008; Saarela et al. 2010; Minaya et al. 2013), hence their current recognition in monotypic subtribes. Within Anthoxanthinae, a molecular study has demonstrated the need for recognition of one or three genera (Pimentel et al. 2013). The shared spikelet characteristics of Anthoxanthinae and Phalaridinae may be plesiomorphies or may have evolved in parallel. Some studies represented Phalaridinae solely by Anthoxanthum (e.g. Soreng and Davis 2000), which in retrospect is uninformative in shedding light on the origins of Phalaridinae as now understood (since Anthoxanthum is part of a separate lineage). Torreyochloinae includes two genera, Amphibromus and Torreyochloa, a circumscription based on plastid and nuclear ribosomal phylogenies (Soreng and Davis 2000; Davis and Soreng 2007; Soreng et al. 2007; Saarela et al. 2010). One set of studies that sampled all three genera of Phalaridinae and Torreyochloinae was based on restriction site and morphological characters and found Phalaris, Amphibromus and Torreyochloa to be part of a clade of Aveneae taxa, but none of these three taxa was closely related (Soreng and Davis 1998, 2000). In Saarela et al. (2010)—so far the only study based on DNA sequence data to sample both subtribes and all three genera—all deep branches in plastid and nuclear ribosomal trees were weakly supported and/or unresolved. Although not incongruent with our current results, the unsupported trees in Saarela et al. (2010) do not provide support for the lineage.

Numerous studies have sampled either Phalaridinae or Torreyochloinae. In those that sampled only Phalaris, its placement was either unresolved (Döring et al. 2007) or variously inferred to be the sister group of the rest of Poeae clade 1 (Quintanar et al. 2007; Bouchenak-Khelladi et al. 2008, 2010; Gillespie et al. 2008; Schaefer et al. 2011), Agrostidinae (Minaya et al. 2013) or Aveninae (Grass Phylogeny Working Group II 2012). Studies that sampled only Torreyochloinae identified it as the sister group of the rest of Poeae clade 1 (Soreng and Davis 2000; Davis and Soreng 2007; Soreng et al. 2007) or Agrostidinae plus Brizinae s.s. (Davis and Soreng 2010). The studies that placed Phalaridinae or Torreyochloinae as sister to the rest of Poeae clade 1 are consistent with our ML and BI trees, in which the Phalaridinae–Torreyochloinae clade is strongly supported (at least based on complete plastome data) as the sister group of the rest of the clade. However, those previous studies and our ML and BI trees contrast with the topology of our MP trees, in which the Aveninae clade is weakly (complete plastome data) to strongly (protein cds) supported as the sister group of the rest of the clade. This latter set of relationships has not been found in other studies and likely represents systematic error in the MP tree here given the extremely short internal branches and long terminal branches at the base of the clade.

The novel relationship identified between Phalaridinae and Torreyochloinae requires consideration of possible non-molecular synapomorphies for this lineage. In terms of their gross morphologies, Phalaridinae and Torreyochloinae are distinctive, differing in their inflorescence shape, the number and fertility of florets per spikelet, glume length and the presence or absence of an awn. Phalaridinae is characterized by inflorescences of false spikes, spikelets laterally compressed with a single fertile floret and two proximal sterile lemmas that are shorter than the fertile lemma, glumes exceeding the florets, lemmas awnless and calluses glabrous (Voshell et al. 2011). Torreyochloinae is a morphologically heterogeneous subtribe. Torreyochloa is characterized by terminal paniculate inflorescences, spikelets laterally compressed to terete with 2–8 florets, glumes rounded to slightly keeled, unawned and shorter than the lowest lemma, lemmas 5–7-nerved (these prominent and scaberulous) and unawned and calluses glabrous (Watson and Dallwitz 1992 onwards; Davis 2007). It is one of several genera in Poeae clade 1 (also including genera of Brizinae s.l.) with spikelet structure characteristic of the traditional Poeae (i.e. simple spikelets with short glumes, several florets, 3–5-nerved lemmas vs. long glumes, 1–several florets, (3)5–11-nerved lemmas and/or geniculate dorsal awns in traditional Aveneae; Clayton and Renvoize 1986), which are mostly part of Poeae clade 1 (Soreng et al. 2007). Its sister taxon, Amphibromus, is characterized by terminal paniculate inflorescences, spikelets laterally compressed with 2–10(–12) fertile florets, glumes rounded to slightly keeled, unawned and shorter than or subequal to the lowest lemma, lemmas 2–4-toothed with teeth extending into short bristles, lemmas dorsally awned from about the middle and calluses hairy (Watson and Dallwitz 1992 onwards; Weiller et al. 2009). The spikelet structure of Torreyochloa may be plesiomorphic in Poeae, or may be a result of convergent evolution in Poeae clades 1 and 2 (Soreng et al. 1990; Soreng and Davis 2000), but a possible hybrid origin for this taxon should not be ruled out without supporting evidence. We are not able to identify any putative synapomorphies for the Phalaridinae–Torreyochloinae clade.

Poeae clade 2

Poeae clade 2 is strongly supported in our analyses, and comprises two major subclades based on the current sampling. One strongly supported subclade includes Dactylidinae, Holcinae, Loliinae and Airinae, and a sister relationship between Dactylidinae and Loliinae is supported in most analyses. In the ML and BI analyses, Airinae and Holcinae are identified as the successive sister groups of the rest of the subclade, whereas in one MP tree (complete data) these taxa are recovered as clade but with no support (BP <50), and in the other (protein cds) their relationships are unresolved. As in Poeae clade 1, these discordant topologies are likely a function of the very short branches at the base of the subclade. Recovery of this four-tribe subclade is consistent with previous studies that identified a larger clade comprising these subtribes as well as Ammochloinae, Cynosurinae, Parapholiinae and Sesleriinae, based on one to three plastid regions (Catalán et al. 2004; Davis and Soreng 2007; Döring et al. 2007; Quintanar et al. 2007; Soreng et al. 2007; Bouchenak-Khelladi et al. 2008; Schneider et al. 2009, 2012; Schaefer et al. 2011; Grass Phylogeny Working Group II 2012) and combined plastid (four regions) and nuclear (ITS and β-amylase) data (Minaya et al. 2013).

To properly interpret the results of earlier studies in the context of the current subtribal classification of Poeae, the recently revised nomenclature of some previously sampled taxa in light of knowledge of their evolutionary histories must be considered. Some species treated as Avenula (A. albinervis, A. compressa, A. gervaisii, A. hookeri, A. pratensis, A. sulcata) and Helictotrichon (H. bromoides, H. schellianum) are now recognized in the genus Helictochloa (Romero-Zarco 2011), which is included in Airinae, and Deschampsia flexuosa is now recognized in Avenella (Chiapella 2007), also included in Airinae. Only Avenula pubescens remains in Avenula s.s. (Romero-Zarco 2011), and it combines characters of Helictotrichon and Helictochloa, but its plastid is phylogenetically isolated from any of the above, possibly aligning with Poinae s.l. Soreng and Davis (2000), as yet unaware of the nrDNA problem, commented on the intermediate nature of Avenula pubescens between Avenula subgen. Pratavenastrum (now Helictochloa) and Helictotrichon. Helictochloa and Deschampsia are apparently part of a set of taxa (also including Avenella, Sesleria, Scolochloa and others) possibly derived from reticulation between plastid-based Poeae clade 1 and Poeae clade 2, or involved in that event or events. They and others align with traditional Poeae in plastid analysis (i.e. Poeae clade 2) and are intermediate between those and taxa with traditional Aveneae-type plastids (i.e. Poeae clade 1) in nrDNA trees (in nrDNA trees the Aveneae taxa arise from within Poeae, and the placements of various genera are quite incongruent) (e.g. Quintanar et al. 2007). This is a complex area that needs further study.

In earlier plastid trees, relationships among the lineages in the subclade were unresolved and/or weakly supported, with the exception of a sister group relationship inferred between Cynosurinae and Parapholiinae (Davis and Soreng 2007; Soreng and Gillespie 2007; Schneider et al. 2012; Minaya et al. 2013), and are not in conflict with the set of relationships inferred here. The combined nuclear and plastid tree in Minaya et al. (2013) is much better resolved, identifying the following strongly supported topology for this subclade: {Airinae (Helictochloa bromoides, as Avenula bromoides) [Dactylidinae (Cynosurinae, Parapholiinae)]}, and identifying a second clade comprising Holcinae, Airinae (Avenella flexuosa, as Deschampsia flexuosa, and Corynephorus) and Sesleriinae. Sessleriinae falls within Poeae clade 2 in plastid analyses, but is nested within Poeae clade 1 taxa in nrDNA analyses. Sesleria has a strange morphology, including a bract below the inflorescence and oddly shaped spikelets, that are suggestive of some disruption of the developmental patterns, possibly resulting from its reticulate origin (R. J. Soreng, pers. obs). The placement of Dactylidinae in the combined nuclear and plastid tree of Minaya et al. (2013) contrasts with its strongly supported placements as sister to Loliinae in our ML and BI trees based on the complete plastome data. We are not able to compare our plastome tree with their plastid tree, as it was not shown.

In earlier studies, two subtribes in the subclade were not monophyletic with respect to their current circumscriptions: Airinae, with Aira, Avenella, Corynephorus and Periballia comprising a lineage separate from Helictochloa (Davis and Soreng 2007; Döring et al. 2007; Quintanar et al. 2007; Soreng et al. 2007; Saarela et al. 2010; Schaefer et al. 2011; Minaya et al. 2013); and Holcinae, with a HolcusVahlodea clade and Deschampsia s.s. representing separate lineages (Davis and Soreng 2007; Döring et al. 2007; Quintanar et al. 2007; Saarela et al. 2010). We do not yet have sufficient plastome sampling to address the monophyly of these subtribes, having sampled only a single exemplar from each. A first strategy for future work should be to obtain plastomes from the currently unsampled subtribes Ammochloinae, Cynosurinae, Parapholiinae and Sesleriinae, and each of the putative lineages representing the non-monophyletic Airinae and Holcinae (at least two of these are currently unsampled). Further sampling in Helictochloa (some 30 species) and Deschampsia s.s. should aim to maximize the phylogenetic diversity in these genera as characterized in recent studies (Chiapella 2007; Winterfeld et al. 2014).

The second major subclade in Poeae clade 2 comprises representatives of Poinae and Coleanthinae. Like the other subclade, this major lineage has been identified in plastid trees in numerous studies (Davis and Soreng 2007; Gillespie et al. 2007; Quintanar et al. 2007; Soreng et al. 2007; Bouchenak-Khelladi et al. 2008; Schaefer et al. 2011; Grass Phylogeny Working Group II 2012; Schneider et al. 2012). Several subtribes that were recognized recently (Soreng et al. 2007), including Phleinae, Cinninae, Alopecurinae and Beckmanniinae, are now included in a more broadly defined subtribe Poinae on the basis of phylogenetic data, although numerous aspects of deep relationship in this large subtribe remain unresolved for plastid data (Gillespie et al. 2007, 2008, 2010). The maximally supported pattern of relationships in our plastome trees, with Coleanthinae identified as the sister group of a PhleumPoa clade (i.e. Poinae), agrees with the earlier plastid trees of Gillespie et al. (2007, 2008). The affinities of the monotypic subtribe Miliinae, not sampled here, remain somewhat unclear and are discordant in plastid and nuclear ribosomal trees. In plastid trees the tribe lies on a very long branch, is weakly allied with Phleum, and the Phleum–Miliinae lineage is identified as the sister group of Poinae, but with poor support (Gillespie et al. 2008). In contrast, in nuclear ribosomal trees a weakly supported Phleum–Miliinae lineage falls outside the Poinae clade, with unclear affinities to a clade of Poinae taxa excluding Poa, a Poa clade and Coleanthinae (Gillespie et al. 2008, 2010). In combined plastid and nuclear analyses a weakly supported Phleum–Miliinae lineage is the sister group of a Poa clade (Gillespie et al. 2010), but this may not be an accurate reflection of evolutionary history given the discordance between the plastid and nuclear ribosomal data partitions. In a nuclear β-amylase tree, Milium is sister to a clade that includes Poinae plus a paraphyletic Coleanthinae, whereas in a combined nuclear (ITS, β-amylase) and plastid tree Milium is sister to Poinae, and the Milium-Poinae clade is sister to Coleanthinae (Minaya et al. 2013). Although plastome data for Miliinae may help clarify its maternal affinities with respect to Phleum and the rest of the Poinae clade, additional nuclear data will be necessary to reconstruct its possible hybrid origins.

Within Loliinae the plastome data indicate that Lolium is nested within a paraphyletic Schedonorus, that Festuca altissima (also known as Schedonorus altissimus) is more closely related to LoliumSchedonorus (these being part of a ‘broad-leaved’ clade of Festuca s.l.) than F. ovina (‘fine-leaved’ clade), and that recognition of Schedonorus and Lolium as genera renders Festuca paraphyletic. All of these findings agree with the results of previous studies (Torrecilla and Catalán 2002; Catalán et al. 2004; Inda et al. 2008, 2014; Hand et al. 2010, 2013). In Soreng et al. (2014), Schedonorus is treated as a synonym of Lolium, based on the phylogeny in Catalán et al. (2009) and consistent with the plastome phylogeny here. The LoliumSchedonorus clade lies on a fairly long branch, and is further supported by 14 unambiguous indels. Schedonorus arundinaceus is a hexaploid species complex comprising three morphologically and physiologically distinct forms recognized as Continental, Mediterranean and rhizomatous (Hand et al. 2010). Hand et al. (2010) sampled each of these forms and related species in the clade to reconstruct the evolutionary history of this agriculturally important pasture grass, and found each of the forms to have different origins. In their plastid tree, the rhizomatous and Continental forms were part of a clade that is the sister group of the clade including Lolium and S. pratensis, with the Mediterranean form placed elsewhere. Schedonorus arundinaceus was similarly not monophyletic in their nuclear ribosomal or other nuclear trees, and some aspects of their interrelationships differed with respect to the plastid tree.

Two complete plastomes are now available for Schedonorus arundinaceus, one of which was sequenced here. The two accessions of this species analysed in our 46-taxon trees are not resolved as a clade in either of the MP analyses [see Supporting Information], as they are in the ML and BI trees; this is likely a function of the extremely long branch of the accession sequenced by Cahoon et al. (2010), rather than the possibility that the accessions may represent different forms. The previously published plastome and our new one for the species both represent the Contintental form—the most common form in North America. Cahoon et al. (2010) made this identification explicit, as they sampled ‘KY31’ [Kentucky 31], a widespread cultivar of the Continental form. Our plastome is from a field-collected specimen from British Columbia, which we identify as the Continental form based on its lack of rhizomes, the main diagnostic character for differentiating these morphotypes (Hand et al. 2010). This identification is consistent with BLAST comparisons of matK variation among our sample and those of the Continental and rhizomatous forms sequenced by Hand et al. (2010). The matK sequences of these samples differ by one substitution that varies within and among the two forms, compared with the more divergent matK sequences in the Mediterranean form, as illustrated in the matK tree in Hand et al. (2010). Although the BLAST search does not unambiguously identify our sample as either the Contintental or rhizomatous form, it does rule out the possibility of it being the Mediterranean form, at least based on current knowledge of variation in matK in that form.

Hand et al. (2013) included the Schedonorus arundinaceus plastome generated by Cahoon et al. (2010) in an analysis with four Festuca, Lolium and Schedonorus plastomes they generated, and observed considerable divergence among S. arundinaceus and the other Loliinae taxa, as we do here. They suggested that this divergence may be due to sequencing errors in the S. arundinaceus genome, and that additional plastomes should be generated from the species to determine whether the observed variation is real or artefactual. The previously sequenced plastome for S. arundinaceus shows 2.7 % difference from the new plastome from this taxon sequenced here. The top hits from a BLAST query of the previously sequenced regions rpoC2 and ccsA, which contain numerous substitutions in the alignment, show the top hits to be the respective regions from species within the PACMAD clade (data not shown), suggestive of some errors in this genome. Variation in our new plastome from S. arundinaceus is more in line with the other three species in the LoliumSchedonorus clade, further indicating problems with the earlier plastome. Inclusion of the long-branch S. arundinaceus genome differentially affects ML BS for the core Pooideae in the complete and protein-coding analyses compared with the 45-taxon analyses, for reasons that are unclear.

Bromeae–Triticeae

As in previous phylogenies, Bromeae and Triticeae are a strongly supported clade in our analyses, and the clade is further supported by 20 unambiguous indels. We have not sampled Littledalea, which is the sister group of Bromeae–Triticeae. The pattern of relationships among the three sampled Triticeae genera agree with plastid (Petersen and Seberg 1997; Mason-Gamer et al. 2002), nuclear (Mason-Gamer 2005) and combined plastid/nuclear phylogenies (Escobar et al. 2011) of the subtribe. An earlier version of the T. aestivum plastome (Ogihara et al. 2000) was recently found to contain sequence from the rice plastid genome (Bahieldin et al. 2014). This contaminated genome was included in an early version of our plastome matrix, and in our preliminary MP and ML analyses the terminal branch of T. aestivum was considerably longer than those of the other Aegilops and Triticum taxa in the trees, indicative of this error (data not shown). In the current matrix containing the corrected T. aestivum plastome sequence, the terminal branch for this taxon is very similar in length to those of all other taxa in the clade. Plastomes for the Aegilops and other Triticum species (except T. aestivum) were sequenced by Middleton et al. (2014). Their phylogeny depicting relationships between wheat, rye and barley was based on a 37 kb subset (<30 %) of the whole plastome. We included the full plastomes for these species in our analyses, and find the relationships among Triticeae taxa to be maximally supported and identical to those reported by Middleton et al. (2014).

Unique plastome features

Within Pooideae the presence of the rps16-trnQ insertion solely in Brachyelytrum suggests that the loss of this insertion may be synapomorphic for the remaining genera. The insertion in the rps16-trnQ region of Phaenosperma and Melica is of note as, after Brachyelytrum, these two genera are the earliest to diverge in our analyses. A greater level of sampling within Phaenospermateae and Meliceae as well as sampling from the early diverging Nardeae and Lygeeae are needed to clarify the evolutionary history of these indels.

One insertion of mitochondrial origin was identified in the plastome of Triticum monococcum. This is currently the fourth documented case of plastome regions exhibiting mitochondrial homology. This type of gene transfer was first documented by Goremykin et al. (2009) in Daucus carota. The second was located by Straub et al. (2013) in Asclepias and the third was located by Wysocki et al. (2015) in two species within the Parianinae lineage of the Bambusoideae (Eremitis sp., Pariana radiciflora). While the first three documented insertions were extensively tested for erroneous assembly, the T. monococcum plastome was not sequenced by our team and cannot currently be verified.

One insertion of nuclear homology was identified in the plastome of Triticum urartu. The presence of this insertion, as well as the mitochondrial insertion, in the inverted repeat region suggests that it may have been retained due to the conserved nature of this region. This plastome was also not sequenced by our team so the presence of this insertion cannot be verified and may be an artefact of mis-assembly.

Conclusions

Our phylogenomic analysis of whole plastomes resolve relationships at the base of Pooideae that have varied in earlier studies, and provide new insights into several aspects of relationship among tribes of Poeae, including a strongly supported novel relationship between Torreyochloinae and Phalaridinae. Plastomes representing Lygeae, Nardeae, Brylkinieae and Littledalea are needed to complete tribal-level plastome sampling, and several subtribes of Poeae are as yet unsampled. Our results demonstrate that inclusion of non-coding data in whole plastome analyses provides important characters for recovering robust support at deep nodes, compared with protein-coding data alone. Given rapid advances in next-generation sequencing, achieving a densely sampled plastome-based phylogeny of Pooideae is a realistic goal that we are working towards.

Sources of Funding

This work was supported in part by the Plant Molecular Biology Center, the Department of Biological Sciences at Northern Illinois University and the National Science Foundation under grant numbers DEB-1120750 to L.G.C., DEB-1120856 to S.A.K. and DEB-1120761 to M.R.D. and DEB-0830020 to J.I.D. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Contributions by the Authors

M.R.D., S.A.K., L.G.C. and J.I.D. conceived and designed the study. J.C.P., P.P.E. and D.R.M. provided technical support. W.P.W., C.F.B., M.R.D., J.I.D., J.M.S. and R.J.S. acquired the data. W.P.W., M.R.D. and J.M.S. analysed and interpreted the data. J.M.S., W.P.W. and M.R.D. wrote the manuscript. J.M.S., W.P.W., M.R.D., J.I.D., R.J.S., C.F.B., S.A.K. and L.G.C. contributed to revising the manuscript.

Conflict of Interest Statement

None declared.

Supporting Information

The following additional information is available in the online version of this article –

Table S1. Summary of the Pooideae classification of Soreng et al. (2014).

Table S2. List of completed and publicly available whole plastomes in Poaceae (as of 25 September 2014).

Figure S1. ML phylogram of complete plastomes of 45 pooid grasses and one outgroup taxon, including the highly divergent, previously banked sequence of Schedonorus arundinaceus (taxon in bold; GenBank accession NC_011713). Bootstrap values are indicated only when at least one is <100 (ML bootstrap value precedes MP bootstrap value). ML and BI topologies were identical. Posterior probabilities at all nodes = 1.00. ‘NR’ indicates a node not resolved or supported above the 50 % bootstrap level in the MP analysis. In the MP analysis, the boldfaced S. arundinaceus is sister to a clade comprising the other S. arundinaceus sample, Festuca pratensis, Lolium multiflorum and L. perenne; the latter clade receives 71 % BS. Clades marked with a cross are reversed in the MP tree. The MP bootstrap value of the node affected by the reversal of these clades reflects that of the reversed topology. Taxa marked with diamonds are a maximally supported clade in the MP tree.

Figure S2. ML phylogram of the protein cds of 45 pooid grasses and one outgroup taxon, including the highly divergent, previously banked sequence of Schedonorus arundinaceus (taxon in bold). Support values are indicated only when at least one is less than the maximum possible value (ML bootstrap value precedes MP bootstrap value, which precedes posterior probability value). ML and MP topologies were identical. ‘NR’ indicates a node not resolved in the strict consensus of most parsimonious trees or not supported above the 50 % bootstrap level. In the MP analysis, the boldfaced S. arundinaceus is sister to a clade comprising the other S. arundinaceus sample, Festuca pratensis, Lolium multiflorum and L. perenne; the latter clade receives 97 % BS. A star indicates clades that were reversed in the BI tree. The posterior probability of the node affected by the reversal of these clades reflects that of the reversed topology.

Dataset S1. Details of scored indels in the 46-taxon matrix.

Acknowledgements

We thank K. Murrell for bioinformatics and laboratory assistance, Y. Yin for providing computational resources and two anonymous reviewers for constructive feedback on an earlier version of the manuscript.

Literature Cited

Asano
T
,
Tsudzuki
T
,
Takahashi
S
,
Shimada
H
,
Kadowaki
K
.
2004
.
Complete nucleotide sequence of the sugarcane (Saccharum officinarum) chloroplast genome: a comparative analysis of four monocot chloroplast genomes
.
DNA Research
 
11
:
93
99
.
Bahieldin
A
,
Al-Kordy
MA
,
Shokry
AM
,
Gadalla
NO
,
Al-Hejin
AMM
,
Sabir
JSM
,
Hassan
SM
,
Al-Ahmadi
AA
,
Schwarz
EN
,
Eissa
HF
,
El-Domyati
FM
,
Jansen
RK
.
2014
.
Corrected sequence of the wheat plastid genome
.
Comptes Rendus Biologies
 
337
:
499
502
.
Barrett
CF
,
Davis
JI
.
2012
.
The plastid genome of the mycoheterotrophic Corallorhiza striata (Orchidaceae) is in the relatively early stages of degradation
.
American Journal of Botany
 
99
:
1513
1523
.
Barrett
CF
,
Davis
JI
,
Leebens-Mack
J
,
Conran
JG
,
Stevenson
DW
.
2013
.
Plastid genomes and deep relationships among the commelinid monocot angiosperms
.
Cladistics
 
29
:
65
87
.
Barrett
CF
,
Specht
CD
,
Leebens-Mack
J
,
Stevenson
DW
,
Zomlefer
WB
,
Davis
JI
.
2014
.
Resolving ancient radiations: can complete plastid gene sets elucidate deep relationships among the tropical gingers (Zingiberales)?
Annals of Botany
 
113
:
119
133
.
Beard
JB
.
2012
.
Origin, biogeographical migrations and diversifications of turfgrasses
.
Research Report SR132
.
East Lansing
:
Michigan State University
,
1
26
.
Bentham
G
.
1861
.
Flora Hongkongensis: a description of the flowering plants and ferns of the island of Hongkong
 .
London
:
Lovell Reeve
.
Bergsten
J
.
2005
.
A review of long-branch attraction
.
Cladistics
 
21
:
163
193
.
Besnard
G
,
Christin
P-A
,
Malé
P-JG
,
Coissac
E
,
Ralimanana
H
,
Vorontsova
MS
.
2013
.
Phylogenomics and taxonomy of Lecomtelleae (Poaceae), an isolated panicoid lineage from Madagascar
.
Annals of Botany
 
112
:
1057
1066
.
Blaner
A
,
Schneider
J
,
Röser
M
.
2014
.
Phylogenetic relationships in the grass family (Poaceae) based on the nuclear single copy locus topoisomerase 6 compared with chloroplast DNA
.
Systematics and Biodiversity
 
12
:
111
124
.
Bortiri
E
,
Coleman-Derr
D
,
Lazo
GR
,
Anderson
OD
,
Gu
YQ
.
2008
.
The complete chloroplast genome sequence of Brachypodium distachyon: sequence comparison and phylogenetic analysis of eight grass plastomes
.
BMC Research Notes
 
1
:
61
.
Bouchenak-Khelladi
Y
,
Salamin
N
,
Savolainen
V
,
Forest
F
,
van der Bank
M
,
Chase
MW
,
Hodkinson
TR
.
2008
.
Large multi-gene phylogenetic trees of the grasses (Poaceae): progress towards complete tribal and generic level sampling
.
Molecular Phylogenetics and Evolution
 
47
:
488
505
.
Bouchenak-Khelladi
Y
,
Verboom
GA
,
Savolainen
V
,
Hodkinson
TR
.
2010
.
Biogeography of the grasses (Poaceae): a phylogenetic approach to reveal evolutionary history in geographical space and geological time
.
Botanical Journal of the Linnean Society
 
162
:
543
557
.
Burke
SV
,
Grennan
CP
,
Duvall
MR
.
2012
.
Plastome sequences of two New World bamboos—Arundinaria gigantea and Cryptochloa strictiflora (Poaceae)—extend phylogenomic understanding of Bambusoideae
.
American Journal of Botany
 
99
:
1951
1961
.
Burke
SV
,
Clark
LG
,
Triplett
JK
,
Grennan
CP
,
Duvall
MR
.
2014
.
Biogeography and phylogenomics of New World Bambusoideae (Poaceae), revisited
.
American Journal of Botany
 
101
:
886
891
.
Burleigh
JG
,
Bansal
MS
,
Eulenstein
O
,
Hartmann
S
,
Wehe
A
,
Vision
TJ
.
2011
.
Genome-scale phylogenetics: inferring the plant tree of life from 18,896 gene trees
.
Systematic Biology
 
60
:
117
125
.
Cahoon
AB
,
Sharpe
RM
,
Mysayphonh
C
,
Thompson
EJ
,
Ward
AD
,
Lin
A
.
2010
.
The complete chloroplast genome of tall fescue (Lolium arundinaceum; Poaceae) and comparison of whole plastomes from the family Poaceae
.
American Journal of Botany
 
97
:
49
58
.
Calsa Júnior
T
,
Carraro
DM
,
Benatti
MR
,
Barbosa
AC
,
Kitajima
JP
,
Carrer
H
.
2004
.
Structural features and transcript-editing analysis of sugarcane (Saccharum officinarum L.) chloroplast genome
.
Current Genetics
 
46
:
366
373
.
Catalán
P
,
Olmstead
RG
.
2000
.
Phylogenetic reconstruction of the genus Brachypodium P. Beauv. (Poaceae) from combined sequences of chloroplast ndhF gene and nuclear ITS
.
Plant Systematics and Evolution
 
220
:
1
19
.
Catalán
P
,
Kellogg
EA
,
Olmstead
RG
.
1997
.
Phylogeny of Poaceae subfamily Pooideae based on chloroplast ndhf gene sequences
.
Molecular Phylogenetics and Evolution
 
8
:
150
166
.
Catalán
P
,
Torrecilla
P
,
Rodríguez
JÁL
,
Olmstead
RG
.
2004
.
Phylogeny of the festucoid grasses of subtribe Loliinae and allies (Poeae, Pooideae) inferred from ITS and trnL–F sequences
.
Molecular Phylogenetics and Evolution
 
31
:
517
541
.
Catalán
P
,
Soreng
RJ
,
Peterson
PM
.
2009
.
Festuca aloha and F. molokaiensis (Poaceae: Loliinae), two new species from Hawai'i
.
Journal of the Botanical Research Institute of Texas
 
3
:
51
58
.
Catalán
P
,
Müller
J
,
Hasterok
R
,
Jenkins
G
,
Mur
LAJ
,
Langdon
T
,
Betekhtin
A
,
Siwinska
D
,
Pimentel
M
,
López-Alvarez
D
.
2012
.
Evolution and taxonomic split of the model grass Brachypodium distachyon
.
Annals of Botany
 
109
:
385
405
.
Charles
M
,
Tang
H
,
Belcram
H
,
Paterson
A
,
Gornicki
P
,
Chalhoub
B
.
2009
.
Sixty million years in evolution of soft grain trait in grasses: emergence of the softness locus in the common ancestor of Pooideae and Ehrhartoideae, after their divergence from Panicoideae
.
Molecular Biology and Evolution
 
26
:
1651
1661
.
Chiapella
J
.
2007
.
A molecular phylogenetic study of Deschampsia (Poaceae: Aveneae) inferred from nuclear ITS and plastid trnL sequence data: support for the recognition of Avenella and Vahlodea
.
Taxon
 
56
:
55
64
.
Christin
P-A
,
Spriggs
E
,
Osborne
CP
,
Strömberg
CAE
,
Salamin
N
,
Edwards
EJ
.
2014
.
Molecular dating, evolutionary rates, and the age of the grasses
.
Systematic Biology
 
63
:
153
165
.
Clark
LG
,
Zhang
W
,
Wendel
JF
.
1995
.
A phylogeny of the grass family (Poaceae) based on ndhF sequence data
.
Systematic Botany
 
20
:
436
460
.
Clayton
WD
,
Renvoize
SA
.
1986
.
Genera graminum. Grasses of the world
.
Kew Bulletin
 
13
:
1
389
.
Clayton
WD
,
Vorontsova
MS
,
Harman
KT
,
Williamson
H
.
2006 onwards
.
GrassBase—The Online World Grass Flora
. .
Cox
MP
,
Peterson
DA
,
Biggs
PJ
.
2010
.
SolexaQA: at-a-glance quality assessment of Illumina second-generation sequencing data
.
BMC Bioinformatics
 
11
:
485
.
Cronn
R
,
Liston
A
,
Parks
M
,
Gernandt
DS
,
Shen
R
,
Mockler
T
.
2008
.
Multiplex sequencing of plant chloroplast genomes using Solexa sequencing-by-synthesis technology
.
Nucleic Acids Research
 
36
:
e122
.
Cummings
MP
,
King
LM
,
Kellogg
EA
.
1994
.
Slipped strand mispairing in a plastid gene: rpoC2 in grasses (Poaceae)
.
Molecular Biology and Evolution
 
11
:
1
8
.
Darriba
D
,
Taboada
GL
,
Doallo
R
,
Posada
D
.
2012
.
jModelTest 2: more models, new heuristics and parallel computing
.
Nature Methods
 
9
:
772
.
Davis
JI
.
2007
.
Torreyochloa G.L. Church
. In:
Barkworth
ME
,
Capels
KM
,
Long
S
,
Anderton
LK
,
Piep
MB
, eds.
Flora of North America, Volume 24. Magnoliophyta: Commelinidae (in part): Poaceae, part 1
 .
Oxford and New York
:
Oxford University Press
.
Davis
JI
,
Soreng
RJ
.
1993
.
Phylogenetic structure in the grass family (Poaceae) as inferred from chloroplast DNA restriction site variation
.
American Journal of Botany
 
80
:
1444
1454
.
Davis
JI
,
Soreng
RJ
.
2007
.
A preliminary phylogenetic analysis of the grass subfamily Pooideae (Poaceae), with attention to structural features of the plastid and nuclear genomes, including an intron loss in GBSSI
.
Aliso
 
23
:
335
348
.
Davis
JI
,
Soreng
RJ
.
2010
.
Migration of endpoints of two genes relative to boundaries between regions of the plastid genome in the grass family (Poaceae)
.
American Journal of Botany
 
97
:
874
892
.
Davis
JI
,
McNeal
JR
,
Barrett
CF
,
Chase
MW
,
Cohen
JI
,
Duvall
MR
,
Givnish
TJ
,
Graham
SW
,
Petersen
G
,
Pires
JC
.
2013
.
Contrasting patterns of support among plastid genes and genomes for major clades of the monocotyledons
. In:
Wilkin
P
,
Mayo
SJ
, eds.
Early events in monocot evolution
 .
Systematics Association Special Volume Series
.
New York: Cambridge University Press
,
315
349
.
Diekmann
K
,
Hodkinson
TR
,
Wolfe
KH
,
van den Bekerom
R
,
Dix
PJ
,
Barth
S
.
2009
.
Complete chloroplast genome sequence of a major allogamous forage species, Perennial Ryegrass (Lolium perenne L.)
.
DNA Research
 
16
:
165
176
.
Döring
E
,
Schneider
J
,
Hilu
KW
,
Röser
M
.
2007
.
Phylogenetic relationships in the Aveneae/Poeae complex (Pooideae, Poaceae)
.
Kew Bulletin
 
62
:
407
424
.
Duvall
MR
,
Davis
JI
,
Clark
LG
,
Noll
JD
,
Goldman
DH
,
Sanchez-Ken
JG
.
2007
.
Phylogeny of the grasses (Poaceae) revisited
.
Aliso
 
23
:
237
247
.
Escobar
JS
,
Scornavacca
C
,
Cenci
A
,
Guilhaumon
C
,
Santoni
S
,
Douzery
EJP
,
Ranwez
V
,
Glémin
S
,
David
J
.
2011
.
Multigenic phylogeny and analysis of tree incongruences in Triticeae (Poaceae)
.
BMC Evolutionary Biology
 
11
:
181
.
Eserman
LA
,
Tiley
GP
,
Jarret
RL
,
Leebens-Mack
JH
,
Miller
RE
.
2014
.
Phylogenetics and diversification of morning glories (tribe Ipomoeeae, Convolvulaceae) based on whole plastome sequences
.
American Journal of Botany
 
101
:
92
103
.
Essi
L
,
Longhi-Wagner
HM
,
de Souza-Chies
TT
.
2008
.
Phylogenetic analysis of the Briza complex (Poaceae)
.
Molecular Phylogenetics and Evolution
 
47
:
1018
1029
.
Essi
L
,
Wagner
HML
,
de Souza Chies
TT
.
2011
.
New combinations within the Briza complex (Poaceae, Pooideae, Poeae)
.
Novon
 
21
:
326
330
.
Felsenstein
J
.
2005
.
PHYLIP (phylogeny inference package)
 .
Distributed by the author
.
Seattle
:
Department of Genome Sciences, University of Washington
.
Gao
J
,
Gao
LZ
.
2014
.
The complete chloroplast genome sequence of the Phyllostachys sulphurea (Poaceae: Bambusoideae)
.
Mitochondrial DNA
 . .
Gaut
B
,
Yang
L
,
Takuno
S
,
Eguiarte
LE
.
2011
.
The patterns and causes of variation in plant nucleotide substitution rates
.
Annual Review of Ecology, Evolution, and Systematics
 
42
:
245
266
.
Gillespie
LJ
,
Archambault
A
,
Soreng
RJ
.
2007
.
Phylogeny of Poa (Poaceae) based on trnT-trnF sequence data: major clades and basal relationships
.
Aliso
 
23
:
420
434
.
Gillespie
LJ
,
Soreng
RJ
,
Bull
RD
,
Jacobs
SWL
,
Refulio-Rodriguez
NF
.
2008
.
Phylogenetic relationships in subtribe Poinae (Poaceae, Poeae) based on nuclear ITS and plastid trnT-trnL-trnF sequences
.
Botany-Botanique
 
86
:
938
967
.
Gillespie
LJ
,
Soreng
RJ
,
Jacobs
SWL
.
2009
.
Phylogenetic relationships of Australian Poa (Poaceae: Poinae), including molecular evidence for two new genera, Saxipoa and Sylvipoa
.
Australian Systematic Botany
 
22
:
413
436
.
Gillespie
LM
,
Soreng
RJ
,
Paradis
LM
,
Bull
RD
.
2010
.
Phylogeny and reticulation in Poinae subtribal complex based on nrITS, ETS and trnTLF data
. In:
Seberg
O
,
Petersen
G
,
Barfod
A
,
Davis
JI
, eds.
Diversity, phylogeny, and evolution in the monocotyledons
 .
Denmark
:
Aarhus University Press
,
589
617
.
Givnish
TJ
,
Ames
M
,
McNeal
JR
,
McKain
MR
,
Steele
PR
,
dePamphilis
CW
,
Graham
SW
,
Pires
JC
,
Stevenson
DW
,
Zomlefer
WB
,
Briggs
BG
,
Duvall
MR
,
Moore
MJ
,
Heaney
JM
,
Soltis
DE
,
Soltis
PS
,
Thiele
K
,
Leebens-Mack
JH
.
2010
.
Assembling the tree of the monocotyledons: plastome sequence phylogeny and evolution of Poales
.
Annals of the Missouri Botanical Garden
 
97
:
584
616
.
Goremykin
VV
,
Salamini
F
,
Velasco
R
,
Viola
R
.
2009
.
Mitochondrial DNA of Vitis vinifera and the issue of rampant horizontal gene transfer
.
Molecular Biology and Evolution
 
26
:
99
110
.
Gornicki
P
,
Zhu
H
,
Wang
J
,
Challa
GS
,
Zhang
Z
,
Gill
BS
,
Li
W
.
2014
.
The chloroplast view of the evolution of polyploid wheat
.
New Phytologist
 
204
:
704
714
.
Grass Phylogeny Working Group
.
2001
.
Phylogeny and subfamilial classification of the grasses (Poaceae)
.
Annals of the Missouri Botanical Garden
 
88
:
373
457
.
Grass Phylogeny Working Group II
.
2012
.
New grass phylogeny resolves deep evolutionary relationships and discovers C4 origins
.
New Phytologist
 
193
:
304
312
.
Grebenstein
B
,
Röser
M
,
Sauer
W
,
Hemleben
V
.
1998
.
Molecular phylogenetic relationships in Aveneae (Poaceae) species and other grasses as inferred from ITS1 and ITS2 rDNA sequences
.
Plant Systematics and Evolution
 
213
:
233
250
.
Guindon
S
,
Gascuel
O
.
2003
.
A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood
.
Systematic Biology
 
52
:
696
704
.
Hand
ML
,
Cogan
NOI
,
Stewart
AV
,
Forster
JW
.
2010
.
Evolutionary history of tall fescue morphotypes inferred from molecular phylogenetics of the Lolium-Festuca species complex
.
BMC Evolutionary Biology
 
10
:
303
.
Hand
ML
,
Spangenberg
GC
,
Forster
JW
,
Cogan
NOI
.
2013
.
Plastome sequence determination and comparative analysis for members of the Lolium-Festuca Grass species complex
.
G3: Genes|Genomes|Genetics
 
3
:
607
616
.
Hilu
KW
,
Alice
LA
.
1999
.
Evolutionary implications of matK indels in Poaceae
.
American Journal of Botany
 
86
:
1735
1741
.
Hilu
KW
,
Alice
LA
,
Liang
H
.
1999
.
Phylogeny of Poaceae inferred from matK sequences
.
Annals of the Missouri Botanical Garden
 
86
:
835
851
.
Hiratsuka
J
,
Shimada
H
,
Whittier
R
,
Ishibashi
T
,
Sakamoto
M
,
Mori
M
,
Kondo
C
,
Honji
Y
,
Sun
CR
,
Meng
BY
,
Li
YQ
,
Kanno
A
,
Nishizawa
Y
,
Hirai
A
,
Shinozaki
K
,
Sugiura
M
.
1989
.
The complete sequence of the rice (Oryza sativa) chloroplast genome: intermolecular recombination between distinct tRNA genes accounts for a major plastid DNA inversion during the evolution of the cereals
.
Molecular and General Genetics
 
217
:
185
194
.
Hoffmann
MH
,
Schneider
J
,
Hase
P
,
Röser
M
.
2013
.
Rapid and recent world-wide diversification of bluegrasses (Poa, Poaceae) and related genera
.
PLoS ONE
 
8
:
e60061
.
Hsiao
C
,
Jacobs
SWL
,
Chatterton
NJ
,
Asay
KH
.
1998
.
A molecular phylogeny of the grass family (Poaceae) based on the sequences of nuclear ribosomal DNA (ITS)
.
Australian Systematic Botany
 
11
:
667
688
.
Huang
DI
,
Hefer
CA
,
Kolosova
N
,
Douglas
CJ
,
Cronk
QCB
.
2014
.
Whole plastome sequencing reveals deep plastid divergence and cytonuclear discordance between closely related balsam poplars, Populus balsamifera and P. trichocarpa (Salicaceae)
.
New Phytologist
 
204
:
693
703
.
Inda
LA
,
Segarra-Moragues
JG
,
Müller
J
,
Peterson
PM
,
Catalán
P
.
2008
.
Dated historical biogeography of the temperate Loliinae (Poaceae, Pooideae) grasses in the northern and southern hemispheres
.
Molecular Phylogenetics and Evolution
 
46
:
932
957
.
Inda
LA
,
Sanmartín
I
,
Buerki
S
,
Catalán
P
.
2014
.
Mediterranean origin and Miocene–Holocene Old World diversification of meadow fescues and ryegrasses (Festuca subgenus Schedonorus and Lolium)
.
Journal of Biogeography
 
41
:
600
614
.
Jansen
RK
,
Ruhlman
TA
.
2012
.
Plastid genomes of seed plants
. In:
Bock
R
,
Knoop
V
, eds.
Genomics of chloroplasts and mitochondria
 .
The Netherlands
:
Springer
,
103
126
.
Jansen
RK
,
Raubeson
LA
,
Boore
JL
,
dePamphilis
CW
,
Chumley
TW
,
Haberle
RC
,
Wyman
SK
,
Alverson
AJ
,
Peery
R
,
Herman
SJ
,
Fourcade
HM
,
Kuehl
JV
,
McNeal
JR
,
Leebens-Mack
J
,
Cui
L
.
2005
.
Methods for obtaining and analyzing whole chloroplast genome sequences
.
Methods in Enzymology
 
395
:
348
384
.
Jones
SS
,
Burke
SV
,
Duvall
MR
.
2014
.
Phylogenomics, molecular evolution, and estimated ages of lineages from the deep phylogeny of Poaceae
.
Plant Systematics and Evolution
 
300
:
1421
1436
.
Katoh
K
,
Kuma
K-i
,
Toh
H
,
Miyata
T
.
2005
.
MAFFT version 5: improvement in accuracy of multiple sequence alignment
.
Nucleic Acids Research
 
33
:
511
518
.
Kellogg
EA
,
Campbell
CS
.
1987
.
Phylogenetic analyses of the Gramineae
. In:
Soderstrom
TR
,
Hilu
KW
,
Campbell
CS
,
Barkworth
ME
, eds.
Grass systematics and evolution
 .
Washington, DC
:
Smithsonian Institution Press
,
310
322
.
Kim
JS
,
Kim
J-H
.
2013
.
Comparative genome analysis and phylogenetic relationship of order Liliales insight from the complete plastid genome sequences of two lilies (Lilium longiflorum and Alstroemeria aurea)
.
PLoS ONE
 
8
:
e68180
.
Lee
J
,
Kang
Y
,
Shin
SC
,
Park
H
,
Lee
H
.
2014
.
Combined analysis of the chloroplast genome and transcriptome of the Antarctic vascular plant Deschampsia antarctica Desv
.
PLoS ONE
 
9
:
e92501
.
Leseberg
CH
,
Duvall
MR
.
2009
.
The complete chloroplast genome of Coix lacryma-jobi and a comparative molecular evolutionary analysis of plastomes in cereals
.
Journal of Molecular Evolution
 
69
:
311
318
.
Lin
Z
,
Wang
Y
,
Fu
F
,
Ye
C
,
Fan
L
.
2014
.
Complete chloroplast genome of Dongxiang wild rice and its application in phylogenetic analysis
.
Journal of Zhejiang University
 
40
:
397
403
.
Ma
P-F
,
Zhang
Y-X
,
Zeng
C-X
,
Guo
Z-H
,
Li
D-Z
.
2014
.
Chloroplast phylogenomic analyses resolve deep-level relationships of an intractable bamboo tribe Arundinarieae (Poaceae)
.
Systematic Biology
 
63
:
933
950
.
Maier
RM
,
Neckermann
K
,
Igloi
GL
,
Kössel
H
.
1995
.
Complete sequence of the maize chloroplast genome: gene content, hotspots of divergence and fine tuning of genetic information by transcript editing
.
Journal of Molecular Biology
 
251
:
614
628
.
Martin
G
,
Baurens
F-C
,
Cardi
C
,
Aury
J-M
,
D'Hont
A
.
2013
.
The complete chloroplast genome of banana (Musa acuminata, Zingiberales): insight into plastid monocotyledon evolution
.
PLoS ONE
 
8
:
e67350
.
Mason-Gamer
RJ
.
2005
.
The β-amylase genes of grasses and a phylogenetic analysis of the Triticeae (Poaceae)
.
American Journal of Botany
 
92
:
1045
1058
.
Mason-Gamer
RJ
,
Orme
NL
,
Anderson
CM
.
2002
.
Phylogenetic analysis of North American Elymus and the monogenomic Triticeae (Poaceae) using three chloroplast DNA data sets
.
Genome
 
45
:
991
1002
.
Mathews
S
,
Tsai
RC
,
Kellogg
EA
.
2000
.
Phylogenetic structure in the grass family (Poaceae): evidence from the nuclear gene phytochrome B
.
American Journal of Botany
 
87
:
96
107
.
Middleton
CP
,
Senerchia
N
,
Stein
N
,
Akhunov
ED
,
Keller
B
,
Wicker
T
,
Kilian
B
.
2014
.
Sequencing of chloroplast genomes from wheat, barley, rye and their relatives provides a detailed insight into the evolution of the Triticeae tribe
.
PLoS ONE
 
9
:
e85761
.
Minaya
M
,
Pimentel
M
,
Mason-Gamer
R
,
Catalán
P
.
2013
.
Distribution and evolutionary dynamics of Stowaway Miniature Inverted repeat Transposable Elements (MITEs) in grasses
.
Molecular Phylogenetics and Evolution
 
68
:
106
118
.
Mochida
K
,
Shinozaki
K
.
2013
.
Unlocking Triticeae genomics to sustainably feed the future
.
Plant and Cell Physiology
 
54
:
1931
1950
.
Moore
MJ
,
Dhingra
A
,
Soltis
PS
,
Shaw
R
,
Farmerie
WG
,
Folta
KM
,
Soltis
DE
.
2006
.
Rapid and accurate pyrosequencing of angiosperm plastid genomes
.
BMC Plant Biology
 
6
:
17
.
Morris
LM
,
Duvall
MR
.
2010
.
The chloroplast genome of Anomochloa marantoidea (Anomochlooideae; Poaceae) comprises a mixture of grass-like and unique features
.
American Journal of Botany
 
97
:
620
627
.
Nadot
S
,
Bajon
R
,
Lejeune
B
.
1994
.
The chloroplast gene rps4 as a tool for the study of Poaceae phylogeny
.
Plant Systematics and Evolution
 
191
:
27
38
.
Ogihara
Y
,
Isono
K
,
Kojima
T
,
Endo
A
,
Hanaoka
M
,
Shiina
T
,
Terachi
T
,
Utsugi
S
,
Murata
M
,
Mori
N
,
Takumi
S
,
Ikeo
K
,
Gojobori
T
,
Murai
R
,
Murai
K
,
Matsuoka
Y
,
Ohnishi
Y
,
Tajiri
H
,
Tsunewaki
K
.
2000
.
Chinese spring wheat (Triticum aestivum L.) chloroplast genome: complete sequence and contig clones
.
Plant Molecular Biology Reporter
 
18
:
243
253
.
Parks
M
,
Cronn
R
,
Liston
A
.
2009
.
Increasing phylogenetic resolution at low taxonomic levels using massively parallel sequencing of chloroplast genomes
.
BMC Biology
 
7
:
84
.
Peng
Z
,
Lu
T
,
Li
L
,
Liu
X
,
Gao
Z
,
Hu
T
,
Yang
X
,
Feng
Q
,
Guan
J
,
Weng
Q
,
Fan
D
,
Zhu
C
,
Lu
Y
,
Han
B
,
Jiang
Z
.
2010
.
Genome-wide characterization of the biggest grass, bamboo, based on 10,608 putative full-length cDNA sequences
.
BMC Plant Biology
 
10
:
116
.
Peredo
EL
,
Les
DH
,
King
UM
,
Benoit
LK
.
2012
.
Extreme conservation of the psaA/psaB intercistronic spacer reveals a translational motif coincident with the evolution of land plants
.
Journal of Molecular Evolution
 
75
:
184
197
.
Petersen
G
,
Seberg
O
.
1997
.
Phylogenetic analysis of the Triticeae (Poaceae) based on rpoA sequence data
.
Molecular Phylogenetics and Evolution
 
7
:
217
230
.
Pimentel
M
,
Sahuquillo
E
,
Torrecilla
Z
,
Popp
M
,
Catalán
P
,
Brochmann
C
.
2013
.
Hybridization and long-distance colonization at different time scales: towards resolution of long-term controversies in the sweet vernal grasses (Anthoxanthum)
.
Annals of Botany
 
112
:
1015
1030
.
Quintanar
A
,
Castroviejo
S
,
Catalán
P
.
2007
.
Phylogeny of the tribe Aveneae (Pooideae, Poaceae) inferred from plastid trnT-F and nuclear ITS sequences
.
American Journal of Botany
 
94
:
1554
1569
.
Quintanar
A
,
Catalán
P
,
Castroviejo
S
.
2010
.
A review of the systematics and phylogenetics of the Koeleriinae (Poaceae: Poeae)
. In:
Seberg
O
,
Petersen
G
,
Barfod
A
,
Davis
JI
, eds.
Diversity, phylogeny, and evolution in the monocotyledons
 .
Denmark
:
Aarhus University Press
,
539
556
.
Refulio-Rodriguez
NF
.
2007
.
Systematics of Dissanthelium (Poaceae: Pooideae)
.
Doctor of Philosophy
,
The Claremont Graduate University
,
Claremont, CA
.
Rodionov
AV
,
Tyupa
NB
,
Kim
ES
,
Machs
EM
,
Loskutov
IG
.
2005
.
Genomic configuration of the autotetraploid oat species Avena macrostachya inferred from comparative analysis of ITS1 and ITS2 sequences: on the oat karyotype evolution during the early events of the Avena species divergence
.
Russian Journal of Genetics
 
41
:
518
528
.
Rodionov
AV
,
Kotsinyan
AR
,
Gnutikov
AA
,
Dobroradova
MA
,
Machs
EM
.
2013
.
Variability of the ITS1-5.8S rDNA-ITS2 sequence during the divergence of sweet-grass species (Glyceria R. Br.)
.
Russian Journal of Genetics: Applied Research
 
3
:
83
90
.
Romaschenko
K
,
Peterson
PM
,
Soreng
RJ
,
Garcia-Jacas
N
,
Susanna
A
.
2010
.
Phylogenetics of Stipeae (Poaceae: Pooideae) based on plastid and nuclear DNA sequences
. In:
Seberg
O
,
Petersen
G
,
Barfod
A
,
Davis
JI
, eds.
Diversity, phylogeny, and evolution in the monocotyledons
 .
Denmark
:
Aarhus University Press
,
522
537
.
Romaschenko
K
,
Peterson
PM
,
Soreng
RJ
,
Garcia-Jacas
N
,
Futorna
O
,
Susanna
A
.
2012
.
Systematics and evolution of the needle grasses (Poaceae: Pooideae: Stipeae) based on analysis of multiple chloroplast loci, ITS, and lemma micromorphology
.
Taxon
 
61
:
18
44
.
Romaschenko
K
,
Garcia-Jacas
N
,
Peterson
PM
,
Soreng
RJ
,
Vilatersana
R
,
Susanna
A
.
2014
.
Miocene-Pliocene speciation, introgression, and migration of Patis and Ptilagrostis (Poaceae: Stipeae)
.
Molecular Phylogenetics and Evolution
 
70
:
244
259
.
Romero-Zarco
C
.
2011
.
Helictochloa Romero Zarco (Poaceae), a new genus of oat grass
.
Candollea
 
66
:
87
103
.
Ronquist
F
,
Huelsenbeck
JP
.
2003
.
MrBayes 3: Bayesian phylogenetic inference under mixed models
.
Bioinformatics
 
19
:
1572
1574
.
Röser
M
,
Winterfeld
G
,
Grebenstein
B
,
Hemleben
V
.
2001
.
Molecular diversity and physical mapping of 5S rDNA in wild and cultivated oat grasses (Poaceae: Aveneae)
.
Molecular Phylogenetics and Evolution
 
21
:
198
217
.
Ruhfel
BR
,
Gitzendanner
MA
,
Soltis
PS
,
Soltis
DE
,
Burleigh
J
.
2014
.
From algae to angiosperms-inferring the phylogeny of green plants (Viridiplantae) from 360 plastid genomes
.
BMC Evolutionary Biology
 
14
:
23
.
Saarela
JM
,
Graham
SW
.
2010
.
Inference of phylogenetic relationships among the subfamilies of grasses (Poaceae: Poales) using meso-scale exemplar-based sampling of the plastid genome
.
Botany
 
88
:
65
84
.
Saarela
JM
,
Peterson
PM
,
Soreng
RJ
,
Chapman
RE
.
2003
.
A taxonomic revision of the eastern North American and eastern Asian disjunct genus Brachyelytrum (Poaceae): evidence from morphology, phytogeography and AFLPs
.
Systematic Botany
 
28
:
674
692
.
Saarela
JM
,
Liu
Q
,
Peterson
PM
,
Soreng
RJ
,
Paszko
B
.
2010
.
Phylogenetics of the grass ‘Aveneae-type plastid DNA clade’ (Poaceae: Pooideae, Poeae) based on plastid and nuclear ribosomal DNA sequence data
. In:
Seberg
O
,
Petersen
G
,
Barfod
A
,
Davis
JI
, eds.
Diversity, phylogeny, and evolution in the monocotyledons
 .
Denmark
:
Aarhus University Press
,
557
587
.
Sanchez-Ken
G
,
Clark
LG
.
2007
.
Phylogenetic relationships within the Centothecoideae + Panicoideae clade (Poaceae) based on ndhF and rpl16 intron sequences and structural data
.
Aliso
 
23
:
487
502
.
Sandve
SR
,
Fjellheim
S
.
2010
.
Did gene family expansions during the Eocene–Oligocene boundary climate cooling play a role in Pooideae adaptation to cool climates?
Molecular Ecology
 
19
:
2075
2088
.
Saski
C
,
Lee
S-B
,
Fjellheim
S
,
Guda
C
,
Jansen
RK
,
Luo
H
,
Tomkins
J
,
Rognli
OA
,
Daniell
H
,
Clarke
JL
.
2007
.
Complete chloroplast genome sequences of Hordeum vulgare, Sorghum bicolor and Agrostis stolonifera, and comparative analyses with other grass genomes
.
Theoretical and Applied Genetics
 
115
:
571
590
.
Schaefer
H
,
Hardy
OJ
,
Silva
L
,
Barraclough
TG
,
Savolainen
V
.
2011
.
Testing Darwin's naturalization hypothesis in the Azores
.
Ecology Letters
 
14
:
389
396
.
Schneider
J
,
Doring
E
,
Hilu
KW
,
Röser
M
.
2009
.
Phylogenetic structure of the grass subfamily Pooideae based on comparison of plastid matK gene-3′trnK exon and nuclear ITS sequences
.
Taxon
 
58
:
405
424
.
Schneider
J
,
Winterfeld
G
,
Hoffmann
MH
,
Röser
M
.
2011
.
Duthieeae, a new tribe of grasses (Poaceae) identified among the early diverging lineages of subfamily Pooideae: molecular phylogenetics, morphological delineation, cytogenetics and biogeography
.
Systematics and Biodiversity
 
9
:
27
44
.
Schneider
J
,
Winterfeld
G
,
Röser
M
.
2012
.
Polyphyly of the grass tribe Hainardieae (Poaceae: Pooideae): identification of its different lineages based on molecular phylogenetics, including morphological and cytogenetic characteristics
.
Organisms Diversity and Evolution
 
12
:
113
132
.
Shahid Masood
M
,
Nishikawa
T
,
Fukuoka
S
,
Njenga
PK
,
Tsudzuki
T
,
Kadowaki
K
.
2004
.
The complete nucleotide sequence of wild rice (Oryza nivara) chloroplast genome: first genome wide comparative sequence analysis of wild and cultivated rice
.
Gene
 
340
:
133
139
.
Soltis
DE
,
Soltis
PS
.
2004
.
Amborella not a “basal angiosperm”? Not so fast
.
American Journal of Botany
 
91
:
997
1001
.
Soreng
RJ
,
Davis
JI
.
1998
.
Phylogenetics and character evolution in the grass family (Poaceae): simultaneous analysis of morphological and chloroplast DNA restriction site character sets
.
The Botanical Review
 
64
:
1
85
.
Soreng
RJ
,
Davis
JI
.
2000
.
Phylogenetic structure in Poaceae subfamily Pooideae as inferred from molecular and morphological characters: misclassification versus reticulation
. In:
Jacobs
SWL
,
Everett
J
, eds.
Grasses: systematics and evolution
 .
Collingwood, Victoria, Australia
:
CSIRO
,
61
74
.
Soreng
RJ
,
Gillespie
LJ
.
2007
.
Nicoraepoa (Poaceae, Poeae), a new South American genus based on Poa subg. Andinae, and emendation of Poa sect. Parodiochloa of the sub-Antarctic Islands
.
Annals of the Missouri Botanical Garden
 
94
:
821
849
.
Soreng
RJ
,
Davis
JI
,
Doyle
JJ
.
1990
.
A phylogenetic analysis of chloroplast DNA restriction site variation in Poaceae subfam. Pooideae
.
Plant Systematics and Evolution
 
172
:
83
97
.
Soreng
RJ
,
Peterson
PM
,
Davidse
G
,
Judziewicz
EJ
,
Zuloaga
FO
,
Filgueiras
TS
,
Morrone
O
.
2003
.
Catalogue of New World grasses (Poaceae): IV. Subfamily Pooideae
.
Contributions from the United States National Herbarium
 ,
48
:
1
730
.
Soreng
RJ
,
Davis
JI
,
Voionmaa
MA
.
2007
.
A phylogenetic analysis of Poeae sensu lato based on morphological characters and sequence data from three plastid-encoded genes: evidence for reticulation, and a new classification for the tribe
.
Kew Bulletin
 
62
:
425
454
.
Soreng
RJ
,
Davidse
G
,
Peterson
PM
,
Zuloaga
FO
,
Judziewicz
EJ
,
Filgueiras
TS
,
Morrone
O
,
Romaschenko
K
.
2014
.
A world-wide phylogenetic classification of Poaceae (Gramineae): căo (草), capim, çayır, çimen, darbha, ghaas, ghas, gish, gramas, graminius, gräser, grasses, gyokh, he-ben-ke, hullu, kasa, kusa, nyasi, pastos, pillu, pullu, zlaki, etc
. ).
Stamatakis
A
.
2006
.
RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models
.
Bioinformatics
 
22
:
2688
2690
.
Stefanović
S
,
Rice
DW
,
Palmer
JD
.
2004
.
Long branch attraction, taxon sampling, and the earliest angiosperms: Amborella or monocots?
BMC Evolutionary Biology
 
4
:
35
.
Straub
SCK
,
Cronn
RC
,
Edwards
C
,
Fishbein
M
,
Liston
A
.
2013
.
Horizontal transfer of DNA from the mitochondrial to the plastid genome and its subsequent evolution in milkweeds (Apocynaceae)
.
Genome Biology and Evolution
 
5
:
1872
1885
.
Swofford
DL
.
2003
.
PAUP*: Phylogenetic analysis using parsimony (* and other methods)
 .
Version 4.0b10. Sinauer
.
Tang
J
,
Xia
H
,
Cao
M
,
Zhang
X
,
Zeng
W
,
Hu
S
,
Tong
W
,
Wang
J
,
Yu
J
.
2004
.
A comparison of rice chloroplast genomes
.
Plant Physiology
 
135
:
412
420
.
The International Brachypodium Initiative
.
2010
.
Genome sequencing and analysis of the model grass Brachypodium distachyon
.
Nature
 
463
:
763
768
.
Torrecilla
P
,
Catalán
P
.
2002
.
Phylogeny of broad-leaved and fine-leaved Festuca lineages (Poaceae) based on nuclear ITS sequences
.
Systematic Botany
 
27
:
241
251
.
Torrecilla
P
,
López-Rodríguez
JA
,
Catalán
P
.
2004
.
Phylogenetic relationships of Vulpia and related genera (Poeae, Poaceae) based on analysis of ITS and trnL-F sequences
.
Annals of the Missouri Botanical Garden
 
91
:
124
158
.
Triplett
JK
,
Clark
LG
.
2010
.
Phylogeny of the temperate bamboos (Poaceae: Bambusoideae: Bambuseae) with an emphasis on Arundinaria and allies
.
Systematic Botany
 
35
:
102
120
.
Triplett
JK
,
Clark
LG
,
Fisher
AE
,
Wen
J
.
2014
.
Independent allopolyploidization events preceded speciation in the temperate and tropical woody bamboos
.
New Phytologist
 
204
:
66
73
.
Voshell
SM
,
Baldini
RM
,
Kumar
R
,
Tatalovich
N
,
Hilu
KW
.
2011
.
Canary grasses (Phalaris, Poaceae): molecular phylogenetics, polyploidy and floret evolution
.
Taxon
 
60
:
1306
1316
.
Watson
L
,
Dallwitz
MJ
.
1992 onwards
.
The grass genera of the world: descriptions, illustrations, identification, and information retrieval; including synonyms, morphology, anatomy, physiology, phytochemistry, cytology, classification, pathogens, world and local distribution, and references
 .
Version: 12 August 2014 http://delta-intkey.com (16 September 2014)
.
Weiller
CM
,
Jacobs
SWL
,
Thompson
IR
.
2009
.
Amphibromus
. In:
Wilson
A
, ed.
Flora of Australia. Volume 44A, Poaceae 2
 .
Melbourne
:
ABRS/CSIRO Australia
,
130
140
.
Winterfeld
G
,
Doring
E
,
Röser
M
.
2009a
.
Chromosome evolution in wild oat grasses (Aveneae) revealed by molecular phylogeny
.
Genome
 
52
:
361
380
.
Winterfeld
G
,
Schneider
J
,
Röser
M
.
2009b
.
Allopolyploid origin of Mediterranean species in Helictotrichon (Poaceae) and its consequences for karyotype repatterning and homogenisation of rDNA repeat units
.
Systematics and Biodiversity
 
7
:
277
295
.
Winterfeld
G
,
Schneider
J
,
Perner
K
,
Röser
M
.
2014
.
Polyploidy and hybridization as main factors of speciation: complex reticulate evolution within the grass genus Helictochloa
.
Cytogenetic and Genome Research
 
142
:
204
225
.
Wolny
E
,
Lesniewska
K
,
Hasterok
R
,
Langdon
T
.
2011
.
Compact genomes and complex evolution in the genus Brachypodium
.
Chromosoma
 
120
:
199
212
.
Wu
F-H
,
Kan
D-P
,
Lee
S-B
,
Daniell
H
,
Lee
Y-W
,
Lin
C-C
,
Lin
N-S
,
Lin
C-S
.
2009
.
Complete nucleotide sequence of Dendrocalamus latiflorus and Bambusa oldhamii chloroplast genomes
.
Tree Physiology
 
29
:
847
856
.
Wu
Z-Q
,
Ge
S
.
2012
.
The phylogeny of the BEP clade in grasses revisited: evidence from the whole-genome sequences of chloroplasts
.
Molecular Phylogenetics and Evolution
 
62
:
573
578
.
Wysocki
WP
,
Clark
LG
,
Kelchner
SA
,
Burke
SV
,
Pires
JC
,
Edger
PP
,
Mayfield
DR
,
Triplett
JK
,
Columbus
JT
,
Ingram
AL
,
Duvall
MR
.
2014
.
A multi-step comparison of short-read full plastome sequence assembly methods in grasses
.
Taxon
 
63
:
899
910
.
Wysocki
WP
,
Clark
LG
,
Attigala
L
,
Ruiz-Sanchez
E
,
Duvall
MR
.
2015
.
Evolution of the bamboos (Bambusoideae; Poaceae): a full plastome phylogenomic analysis
.
BMC Evolutionary Biology
 
15
:
50
.
Young
HA
,
Lanzatella
CL
,
Sarath
G
,
Tobias
CM
.
2011
.
Chloroplast genome variation in upland and lowland switchgrass
.
PLoS ONE
 
6
:
e23980
.
Yue
J-X
,
Li
J
,
Wang
D
,
Araki
H
,
Tian
D
,
Yang
S
.
2010
.
Genome-wide investigation reveals high evolutionary rates in annual model plants
.
BMC Plant Biology
 
10
:
242
.
Zeng
C-X
,
Zhang
Y-X
,
Triplett
JK
,
Yang
J-B
,
Li
D-Z
.
2010
.
Large multi-locus plastid phylogeny of the tribe Arundinarieae (Poaceae: Bambusoideae) reveals ten major lineages and low rate of molecular divergence
.
Molecular Phylogenetics and Evolution
 
56
:
821
839
.
Zerbino
DR
,
Birney
E
.
2008
.
Velvet: algorithms for de novo short read assembly using de Bruijn graphs
.
Genome Research
 
18
:
821
829
.
Zhang
W
.
2000
.
Phylogeny of the grass family (Poaceae) from rpl16 intron sequence data
.
Molecular Phylogenetics and Evolution
 
15
:
135
146
.
Zhang
Y-J
,
Ma
P-F
,
Li
D-Z
.
2011
.
High-throughput sequencing of six bamboo chloroplast genomes: phylogenetic implications for temperate woody bamboos (Poaceae: Bambusoideae)
.
PLoS ONE
 
6
:
e20596
.
Zhao
L
,
Zhang
N
,
Ma
P-F
,
Liu
Q
,
Li
D-Z
,
Guo
Z-H
.
2013
.
Phylogenomic analyses of nuclear genes reveal the evolutionary relationships within the BEP clade and the evidence of positive selection in Poaceae
.
PLoS ONE
 
8
:
e64642
.

Author notes

Associate Editor: Chelsea D. Specht
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

Comments

0 Comments