TY - JOUR
AU - Zizka,, Georg
AB - Abstract Ochagavia (four species) and Fascicularia (one species) form a well-supported clade of the early-diverging Bromelioideae. The two genera are morphologically similar, but they can be easily discerned on the basis of generative characters. Besides the species distributed on the Chilean mainland, the group includes O. elegans, endemic to the Robinson Crusoe Island of the Juan Fernández Islands. In previous molecular phylogenetic studies, O. elegans formed a sister clade to the remainder of Fascicularia and Ochagavia. A phylogenomic approach, including nearly complete and, in five cases, full plastomes (c. 160 kbp) and the nuclear rDNA cistron (c. 6 kbp), and scanning electron microscope (SEM) images of pollen were used to analyse relationships in the Fascicularia-Ochagavia group. Plastome and nuclear trees were largely congruent and supported previous phylogenetic analyses of O. elegans being sister to the remainder of the group. A divergent phylogenetic position was suggested for O. carnea using different organellar trees. SEM analysis of pollen supported the division of Fascicularia and Ochagavia. Evolutionary and taxonomic implications of our results are discussed. Chile, early-diverging Bromelioideae, gene flow, genome skimming, hybridization, plastid DNA, plastome, pollen, rDNA, Robinson Crusoe Island INTRODUCTION Fascicularia Mez and Ochagavia Phil. (Bromeliaceae, Bromelioideae) are endemic to central and southern Chile. Fascicularia, described in 1894 by Mez, comprises one species with two subspecies (Zizka et al., 1999). Ochagavia, erected in 1856 by Philippi, includes four species (Zizka, Trumpler & Zöllner, 2002). With Greigia Regel, they form the south-western border of the distribution of Bromeliaceae and are, among subfamily Bromelioideae, a group of diploid, terrestrial, non-tank-forming species adapted to temperate or Mediterranean climates (Nelson & Zizka, 1997; Gitaí et al., 2014 ). In all recent phylogenetic analyses, Fascicularia and Ochagavia were assigned to the early-diverging lineages of the early-diverging Bromelioideae (Horres et al., 2000, 2007; Schulte, Horres & Zizka, 2005; Schulte & Zizka, 2008; Schulte, Barfuss & Zizka, 2009; Givnish et al., 2011; Silvestro, Zizka & Schulte, 2014). This makes them particularly important for the understanding of the origin and evolution of Bromelioideae, which are characterized by the species radiation in the Brazilian Mata Atlantica and have a diversity of > 964 species (Gouda, Butcher & Gouda; 2018). Taxonomically, the two genera have caused considerable confusion mainly due to the lack of differences in vegetative characters, resulting in a number of synonyms. However, at the flowering stage, Fascicularia and Ochagavia display striking differences. The most notable distinguishing characters are sepal colour (dark blue vs. pink), relative style and stamen length (shorter than the petals vs. stamens and style exserted), sepal shape (retuse or praemorse vs. ± acute), petal appendages (present vs. absent) and pollen morphology (inaperturate vs. monosulcate) (Ehler & Schill, 1973). Both genera have been recently revised, providing enough characters to reliably identify genera and species (Zizka et al., 1999; Zizka et al., 2002). The first attempt to resolve the relationships of the Fascicularia-Ochagavia group using molecular markers was made by random amplification of polymorphic DNA (RAPD) (Zizka et al., 1999), and this supported the distinction of the genera, although the outgroup Greigia sphacelata (Ruiz & Pav.) Regel fell inside the ingroup. Subsequently, several phylogenetic analyses based on Sanger sequencing of one to eight nuclear and plastid markers included species of both genera. In the early and/or less resolved studies, members of Greigia, Bromelia L. and Deinacanthon Mez can be found nested in the Fascicularia-Ochagavia clade (Schulte et al., 2005; Evans et al., 2015). However, later studies with better resolution confirmed the close relationship of Fascicularia and Ochagavia, together forming a clade among the early-diverging Bromelioideae in a sister position to Greigia and Deinacanthon, and provided a divergence time estimate of 7.5 Mya for the stem and 4.5 Mya for the crown of the Fascicularia-Ochagavia group (Horres et al., 2007; Schulte et al., 2009; Silvestro et al., 2014). Common to most topologies is that the molecular data do not support the delimitation of the two genera as it is suggested by the morphological data. In these cases, O. elegans Phil., a species endemic to the Robinson Crusoe Island (Isla Mas a Tierra) of the Juan Fernández Islands, is sister to a clade of the other Ochagavia spp. investigated and Fascicularia bicolor (Ruiz & Pav.) Mez. Complete separation of Ochagavia and Fascicularia including O. elegans was possible only by using amplified fragment length polymorphisms (AFLPs) (Horres et al., 2007). Here, we present a phylogenomic study of the Fascicularia-Ochagavia group based on nearly complete and full plastome (c. 160 kbp) and nuclear ribosomal DNA (rDNA) cistron (c. 6 kbp) data covering all species and subspecies with mostly two accessions and several outgroup taxa. Using a much larger dataset than in all previous studies, we aim to test the phylogenetic placement of O. elegans and to assess the relationships among other taxa in the Fascicularia-Ochagavia group. Additionally, we include new data on pollen morphology, which has been demonstrated to be particularly useful in taxonomy of Bromelioideae (Halbritter 1992; Leme et al., 2017), and discuss evolutionary, systematic and nomenclatural consequences of our results. MATERIAL AND METHODS Plant material Fifteen samples representing all four Ochagavia spp., both subspecies of F. bicolor, two outgroup taxa of Bromelioideae and the sister subfamily Puyoideae were included in the study. Samples were collected in the field or obtained from the Palmengarten Frankfurt or Utrecht Botanic Gardens. A list of all studied samples with geographical origin, collection history and herbarium voucher information is included in Supporting Information, Table S1. DNA extraction Total genomic DNA was isolated from 23–65 mg of herbarium or silica-gel dried leaves according to the modified CTAB method (Doyle & Doyle, 1990) as described in Horres et al. (2000), including an additional purifying step to precipitate remaining polysaccharides (Michaels, John & Amasino, 1994). The DNA concentration was quantified using a Qubit dsDNA BR Assay Kit and a Qubit 2.0 fluorometer (Invitrogen, Carlsbad, CA, USA). DNA sequencing Illumina paired-end libraries were constructed using the TruSeq compatible genomic DNA library preparation kit for an insert size of 250–450 base pairs (bp). Libraries were multiplexed 96 times (including samples outside of this study) and DNA sequencing with 100-bp paired-end reads was carried out on an Illumina HiSeq 2500 (Illumina, San Diego, CA, USA) platform including positive controls. The construction of sequencing libraries and sequencing was performed by Global Biologics, LLC (Columbia, MO, USA). Raw reads were deposited into the NCBI Sequence Read Archive (SRA) database under the BioProject ID PRJNA556891. Data treatment Raw reads were first filtered to obtain high-quality data by removing adapter sequences and low-quality reads using Trimmomatic v.0.30 (Bolger, Lohse & Usadel, 2014). Bases were discarded in case of quality < Q20 of read ends. The remaining reads were trimmed, if average quality in a 5-bp window was < Q20 and removed if read length fell below 36 bp after trimming. The FASTX‐Toolkit (Gordon & Hannon, 2010) was used to remove duplicate reads. Ribosomal DNA reads were mapped to a reference, which was constructed as follows: quality trimmed reads of Puya mirabilis (Mez) L.B.Sm. S80 were mapped to a 680-bp partial sequence of Puya ferruginea (Ruiz & Pav.) L.B.Sm. (GenBank accession KF265363), using Geneious v.11.1.4 (Kearse et al., 2012), custom sensitivity and five iterations. The resulting consensus sequence was then aligned with the 18S and 26S sequences of Brassica rapa L. subsp. pekinensis (Lour.) Kitam. (GenBank accession KM538956), because these parts of the rDNA cistron could not be assembled using the Puya reads. Finally, the P. mirabilis S80 reads were mapped to this 5874 bp reference, using the Geneious settings as above. The resulting reference sequence was trimmed to 5805 bp representing only the region from the 18S to the 26S rRNA gene, and ITS1 and ITS2 regions were replaced by Ns. Three different plastome (cp) datasets (henceforth called cp-dataset-1, cp-dataset-2 and cp-dataset-3) were generated by read mapping to the Ananas comosus (L.) Merr. plastome (GenBank accession NC026220; Nashima et al., 2015) or assembled de novo. The cp-dataset-1 was generated by read mapping to A. comosus NC026220 containing one inverted repeat (IR) only. The cp-dataset-2 consisted of 86 annotated coding regions. Read mapping was carried out using the Geneious mapper, custom sensitivity and 25 iterations. For cp-dataset-3, full or nearly complete plastomes were de novo assembled using Fast-Plast v.1.2.7 (McKain & Wilson, 2018) under default settings. The pipeline first cleaned the reads using Trimmomatic (Bolger et al., 2014) and extracted plastid-derived reads with Bowtie 2 (Langmead & Salzberg, 2012), using the A. comosus plastome as a reference. Then the reads were de novo assembled by the combination of the De Bruijn graph-based method of SPAdes (Bankevich et al., 2012) with an iterative seed-based microassembly implemented in the custom script afin (https://github.com/afinit/afin/), which tries to close gaps of contigs with low coverage. If no single contig was obtained, SSPACE (Boetzer et al., 2011) was used for scaffolding. The four plastome partitions (LSC, SSC, IRA, IRB; Nashima et al., 2015) were identified, properly oriented and ordered. Finally, coverage analysis was performed using Jellyfish 2 with a threshold of 5× (Marçais & Kingsford, 2011). For plastomes that did not pass the Fast-Plast quality thresholds (ten out of 15), de novo contigs were oriented and aligned with Mulan (Ovcharenko et al., 2005), using the A. comosus reference sequence. Successfully de novo assembled plastomes were annotated using GeSeq (Tillich et al., 2017) with manual adjustments of some start- and stop-codons (ndhF, petB, rbcL rpl2, rpl16, rpl32, rps8). The plastomes were visualized using OGDRAW (Lohse, Drechsel & Bock, 2007). Phylogenetic reconstruction Datasets were separately aligned using MAFFT v.7.388 (Katoh & Standley, 2013) as implemented in Geneious v.11.1.4 using default settings and manually adjusted. In the case of cp-dataset-3, poorly aligned positions and divergent regions of the alignment were eliminated by Gblocks v.0.91b (Castresana, 2000). Phylogenetic relationships were inferred using maximum likelihood (ML) and Bayesian inference (BI). Best-fit nucleotide substitution models were determined for rDNA (introns/exons) and cp-dataset-2 (86 coding regions separately) using ‘greedy’ and ‘rcluster’ algorithms, respectively (Lanfear et al., 2012), RAxML (Stamatakis, 2014) and the corrected Akaike information criterion (AICc) as implemented in PartitionFinder 2 (Lanfear et al., 2017). ML analyses were performed using RAxML graphical front-end raxmlGUI v.1.5 (Silvestro & Michalak, 2012) and GTR+G nucleotide substitution model for both the whole alignment as well as partitioned datasets (rDNA, cp-dataset-2), and they comprised 100 runs and 10 000 thorough bootstrap (BS) pseudoreplicates. BI was conducted in MrBayes v.3.2.6 (Ronquist & Huelsenbeck, 2003) applying GTR, GTR+G and GTR+I+G models for different partitions of the partitioned rDNA and cp-dataset-2 as assigned by PartitionFinder 2 and GTR+G for the non-partitioned cp-datasets. Two independent runs with four Markov chains each (one cold chain and three heated) were carried out for 106 generations sampling trees every 1000th generation. We assessed convergence of the parameters by evaluating the estimated sample size (ESS > 200) in Tracer v.1.7 (Rambaut et al., 2018) and the potential scale reduction factor (PSRF→1) (Gelman & Rubin 1992). The first 25% of the sampled trees were discarded as burn-in, and the two runs were combined. A majority rule consensus of the remaining trees was computed to calculate Bayesian posterior probabilities (PP). Concerning phylogenetic trees, we consider in the following 75–84% BS/0.9–0.94 PP moderate and 85–100% BS/0.95–1 PP high support. To further examine the evolutionary relationships within the species group, a phylogenetic network using the Neighbor-Net algorithm (Bryant & Moulton, 2004), assuming uncorrected p distance and 1000 BS replicates was computed in SplitsTree v.4.14.4 (Huson & Bryant, 2006). Alignments as well as the input and output files, which specify the different data partitioning schemes, are deposited in Zenodo (doi:10.5281/zenodo.3515783). Pollen analysis For scanning electron microscope (SEM) investigations pollen (for sample details, see Supporting Information, Table S1) was placed in small paper pouches and rehydrated for a few seconds to obtain the turgescent pollen state. The material was then dehydrated in 2,2-dimethoxypropane and critical-point dried according to Halbritter (1998). The samples were mounted on stubs with double-sided sticky tape, sputter-coated with gold and investigated using a JEOL JSM-IT300 (Akishima, Japan) scanning microscope. Pollen measurements were carried out on at least five well-preserved pollen grains per studied accession. Descriptive terminology follows Halbritter et al. (2018). RESULTS Data sets Sequencing and high-quality data selection yielded 2.8–8.2 million reads per sample. Of these, 15 303–52 898 reads per sample were mapped to the rDNA reference with 218–761× mean coverage. The final alignment comprised of 5945 bp and all 15 samples. Several samples shared the same rDNA sequences [S81+S83/S84 O. andina (Phil.) Zizka, Trumpler & Zöllner/O. litoralis (Phil.) Zizka, Trumpler & Zöllner; S79+S82 O. elegans, S77+S89 F. bicolor subsp. canaliculata E.C.Nelson & Zizka]. For cp-dataset-1, 154 435 to 1 070 535 reads mapped to the modified Ananas comosus plastome resulting in 167–1610× mean coverage. The final alignment comprised 16 accessions (including Ananas comosus GenBank accession NC026220) and 134 282 bp. For the cp-dataset-2 construction, 224 to 330 936 reads mapped to the particular coding regions ranging from 90 to 6855 bp. Concatenation of the consensus sequences trimmed to the reference resulted in a 79 432-bp alignment. The cp-dataset-3 created with Fast-Plast resulted in fully assembled plastomes of five accessions (F. bicolor subsp. bicolor S78, O. carnea S90, O. elegans S79 and S82 and P. mirabilis S80; see Supporting Information, Table S2). The other accessions had either too low read coverage (i.e. too few plastome-derived reads) or the coverage was uneven across the plastome (i.e. there were regions with coverage < 5× identified during the Fast-Plast run). Our Fast-Plast results clearly showed that the plastome of samples with < c. 120 000 plastome-derived reads cannot be successfully assembled. However, the success cannot be guaranteed even with a higher number of reads (e.g. F. bicolor subsp. canaliculata S89 with 298 734 plastome-derived reads, see Supporting Information, Table S2). De novo assembled plastomes of O. elegans S79 and S82 were identical. After the application of Gblocks, the alignment of cp-dataset-3 resulted in 113 014 bp. Phylogenetic analyses All analyses confirmed the monophyly of the Fascicularia-Ochagavia group. BI and ML phylogenetic trees based on the rDNA dataset showed several moderately to strongly supported clades. The division of the Fascicularia-Ochagavia group from the early-diverging Bromelioideae (Greigia, Bromelia) was moderately (BS) to strongly (PP) supported (Fig. 1). In the Fascicularia-Ochagavia group, O. elegans was in a strongly supported sister position to the remainder of the group. The latter consisted of an unresolved clade of O. andina, O. carnea and O. litoralis in sister position to F. bicolor (moderate PP support). Subspecies of F. bicolor were separated with moderate (BS) to strong (PP) support. Neighbor-Net analysis (Fig. 2) separated the Fascicularia-Ochagavia group from the remainder by a well-supported set of edges. Ochagavia elegans is placed on a long branch, but the number of parallel edges is lower and the path length to the closest congeners is shorter than to Fascicularia or other studied accessions. Figure 1. Open in new tabDownload slide Best scoring ML tree based on the rDNA dataset for Fascicularia-Ochagavia. Numbers at nodes show Bayesian posterior probabilities and bootstrap support percentages (PP/BS). The scale bar below the tree shows the branch length for 0.002 substitutions per nucleotide position. Figure 1. Open in new tabDownload slide Best scoring ML tree based on the rDNA dataset for Fascicularia-Ochagavia. Numbers at nodes show Bayesian posterior probabilities and bootstrap support percentages (PP/BS). The scale bar below the tree shows the branch length for 0.002 substitutions per nucleotide position. Figure 2. Open in new tabDownload slide Split graph resulting from Neighbor-Net analysis using uncorrected p-distances of the rDNA dataset for Fascicularia-Ochagavia. Numbers at the edges indicate bootstrap support. The scale bar indicates genetic distance. Figure 2. Open in new tabDownload slide Split graph resulting from Neighbor-Net analysis using uncorrected p-distances of the rDNA dataset for Fascicularia-Ochagavia. Numbers at the edges indicate bootstrap support. The scale bar indicates genetic distance. The plastome-based analyses provided largely congruent results with the nuclear data. Concerning the cp-dataset-1 (Fig. 3), monophyly of the Fascicularia-Ochagavia group and its sister group relationship with Greigia received maximum support. Ananas and Bromelia were sisters to Greigia and the Fascicularia-Ochagavia group (BS 100, PP 1). In the Fascicularia-Ochagavia group, O. elegans was again in a strongly supported (BS 100, PP 1) sister position to the remainder of the group, in which O. carnea S90 and O. litoralis were sister to O. andina and F. bicolor (BS 100, PP 1). Subspecies of F. bicolor were separated with moderate (BS) to strong (PP) support; O. carnea S85 was placed within Fascicularia as a sister to F. bicolor subsp. bicolor (BS 100, PP 1). The Neighbor-Net analysis of cp-dataset-1 was congruent with the tree analyses (Fig. 4). A strongly supported set of edges separated the Fascicularia-Ochagavia group from the remainder, although O. elegans, placed on long branches, showed a rather distinct position. Similarly, as in the tree analyses, O. carnea S85 grouped with F. bicolor subsp. bicolor. Cp-dataset-3 (de novo plastomes) analyses showed an identical topology (ML/BI, NN; Supporting Information, Fig. S1) as cp-dataset-1, although the separation of subspecies of F. bicolor was not supported by BS and only moderately by PP. The tree topology of the cp-dataset-2 (coding regions) analyses (Supporting Information, Fig. S2) was also largely congruent with previous analyses with the exception of Greigia, which formed a clade with Ananas in a sister position to Bromelia. However, these relationships received no BS support, but strong PP support. Only moderate BS support was assigned for the sister position of the Fascicularia-Ochagavia group with the remainder of Bromelioideae as well as no support for the separation of the subspecies of F. bicolor. Moreover, if partitioning is applied to cp-dataset-2, the Fascicularia-Ochagavia relationship with Greigia/Ananas/Bromelia is unsupported, and O. litoralis S81 and O. andina S83 show up in an unsupported position within Fascicularia (Supporting Information, Fig. S3). Moreover, in partitioned BI analysis of cp-dataset-2, the ESS for tree length of both runs was < 100. Figure 3. Open in new tabDownload slide Best scoring ML tree based on cp-dataset-1 for Fascicularia-Ochagavia. Numbers at nodes show Bayesian posterior probabilities and bootstrap support percentages (PP/BS). The scale bar below the tree shows the branch length for 0.002 substitutions per nucleotide position. Figure 3. Open in new tabDownload slide Best scoring ML tree based on cp-dataset-1 for Fascicularia-Ochagavia. Numbers at nodes show Bayesian posterior probabilities and bootstrap support percentages (PP/BS). The scale bar below the tree shows the branch length for 0.002 substitutions per nucleotide position. Figure 4. Open in new tabDownload slide Split graph resulting from Neighbor-Net analysis using uncorrected p-distances of cp-dataset-1 for Fascicularia-Ochagavia. Numbers at the edges indicate bootstrap support. The scale bar indicates genetic distance. Figure 4. Open in new tabDownload slide Split graph resulting from Neighbor-Net analysis using uncorrected p-distances of cp-dataset-1 for Fascicularia-Ochagavia. Numbers at the edges indicate bootstrap support. The scale bar indicates genetic distance. De novo plastomes Five complete plastomes (F. bicolor subsp. bicolor S78, O. carnea S90, O. elegans S79, S82, P. mirabilis S80) were successfully assembled using Fast-Plast (Supporting Information, Table S2 and Fig. S4). The length of the plastomes varied in size from 158 163 to 162 835 bp. Plastome architecture ranged from 86 453 to 87 800 bp for the large single copy (LSC) region, 18 202 to 18 529 bp for the small single copy (SSC) region, and 26 750 to 26 760 bp for the inverted repeat (IR) region, which are values comparable to the A. comosus plastid genome (GenBank accession NC026220). Overall GC content was conserved, ranging from 37.3 to 37.5%. IR boundaries showed conserved gene content in O. elegans and P. mirabilis. F. bicolor subsp. bicolor S78 revealed additional rps3 and O. carnea S90 additional rps3, rpl14 and rpl16 and a partial copy of rps8 at the IRA-LSC boundary when compared to the remainder (Supporting Information, Fig S4). Concerning the rbcL gene, both accessions of O. elegans showed a premature stop-codon mutation very close to the 3’-end (at 1432 bp) when compared to the remainder. Final plastome sequences are deposited in GenBank (accession numbers MN563795-MN563799, Supporting Information, Table S2). Pollen analysis The pollen of the studied Fascicularia accession (Fig. 5) was more or less spherical and inaperturate. Pollen size was c. 35 µm in diameter. The ornamentation was reticulate with characteristic broad muri and free-standing columellae in the lumina. The pollen of studied Ochagavia (Fig. 6), including O. elegans and O. andina (Supporting Information, Fig. S5), was oblate and monosulcate, with a smooth margin of the sulcus. Pollen diameter was c. 50 µm in the longest axis. The ornamentation was reticulate. The reticulum was heterobrochate with free-standing columellae in the larger lumina. The lumina width decreases towards the sulcus border. Figure 5. Open in new tabDownload slide Pollen of Fascicularia bicolor subsp. bicolor (Zizka 1790). A, B, total view, scale bars 10 µm. C, D, Pollen surface, reticulum with free-standing columellae, scale bars 1 µm. Figure 5. Open in new tabDownload slide Pollen of Fascicularia bicolor subsp. bicolor (Zizka 1790). A, B, total view, scale bars 10 µm. C, D, Pollen surface, reticulum with free-standing columellae, scale bars 1 µm. Figure 6. Open in new tabDownload slide Pollen of Ochagavia carnea (Zizka 1300). A, oblique distal polar view; B, oblique proximal polar view; C, sulcus margin and D, heterobrochate reticulum with some free-standing columellae, scale bars 10 µm. Figure 6. Open in new tabDownload slide Pollen of Ochagavia carnea (Zizka 1300). A, oblique distal polar view; B, oblique proximal polar view; C, sulcus margin and D, heterobrochate reticulum with some free-standing columellae, scale bars 10 µm. DISCUSSION Phylogenetic relationships Our study demonstrates that genome skimming data for Bromeliaceae can yield a set of informative phylogenetic trees based on the whole plastome and nuclear rDNA cistron and provide insights into the evolution of the Fascicularia-Ochagavia group. Reconstructed phylogenetic relationships based on rDNA, cp-dataset-1 and cp-dataset-3 are largely congruent with previous analyses of the Fascicularia-Ochagavia group, although limited sampling of the former studies does not allow a thorough comparison of the topologies. Concerning the Bromelioideae outgroups, plastome-based trees suggested Greigia in a sister position to Fascicularia-Ochagavia, as previously shown by Silvestro et al. (2014). In contrast, in the nuclear tree Greigia grouped with Bromelia, as in ML and BI analyses of Givnish et al., (2011), in which, however, the early-diverging Bromelioideae are under-sampled. Due to limited sampling, we consider our analyses inconclusive on this issue, although the topology of our nuclear tree could be potentially caused by long-branch attraction (e.g. Xu et al., 2017), supported by the fact, that the rDNA dataset did not include Ananas comosus. All but one DNA sequence-based study confirm the monophyly of the group and the sister position of O. elegans to the remainder of the group (Schulte et al., 2005; Horres et al., 2007; Givnish et al., 2011; Silvestro et al., 2014) as reconstructed here. The study by Evans et al. (2015) using three plastid markers revealed a weakly supported clade of O. elegans and Deinacanthon urbanianum (Mez) Mez. However, this relationship might be an artefact of missing Greigia in the analysis. The RAPD-based UPGMA phenogram of Zizka et al. (1999) places G. sphacelata in the Fascicularia-Ochagavia group. Nevertheless, homology among RAPD fragments, especially when comparing rather more distantly related taxa with diverging genome size (Gitaí et al., 2014), has often been questioned (e.g. van de Zande & Bijlsma, 1995). Therefore, we consider the monophyly of Fascicularia-Ochagavia to be conclusively supported by our analyses based on substantial sequence data representing two different genome partitions. Both plastid and nuclear analyses revealed O. elegans in a sister position to the remainder of the Fascicularia-Ochagavia group, which was also reflected by the placement of O. elegans on relatively long branches in the Neighbor-Net analyses. This is in agreement with all previous studies including O. elegans (Zizka et al., 1999; Schulte et al., 2005; Horres et al., 2007; Silvestro et al., 2014), except the AFLP-based analysis of Horres et al. (2007). The separation of O. elegans from a common ancestor was dated to c. 4.5 Mya (95% HPD 2.2–7.3, Silvestro et al., 2014). Ochagavia elegans is endemic to the volcanic Robinson Crusoe Island of the Juan Fernández Islands in the South Pacific, which emerged c. 3.5–4.2 Mya (Stuessy et al., 1984; Baker et al., 1987). Hence, we may assume that the island was colonized once either by the common ancestor or soon after speciation, and O. elegans evolved on Robinson Crusoe Island in isolation until today. Moreover, the rDNA cistron and de novo plastomes of both O. elegans accessions were identical, strongly suggesting that these two accessions are clonal. This is interesting because both accessions have possibly a different collection history. S79 was sampled as a living plant in Palmengarten and originated from RBG Kew (accession 1987–2763), whereas S82 was collected in the Botanical Garden Viña del Mar in 2006. According to Wilkin (1996), vegetative reproduction in O. elegans may well be more prevalent than sexual reproduction, as this species forms dense colonies through vigorous offset production. Therefore, we may assume that the same colony might have been sampled twice in a time span of two decades or more. Alternatively, the intraspecific genetic variation of O. elegans could be exceptionally low, due to e.g. founder effect, low effective population size or, possibly, inbreeding (e.g. Frankham, 1997; Stuessy et al., 2014). However, at this moment we are not able to properly address this issue having only two individuals in our sampling. Except for O. elegans, the phylogenetic relationships in the Fascicularia-Ochagavia group based on information from different genomic partitions are not fully congruent. On one hand, rDNA suggests a sister relationship of Fascicularia to the remaining Ochagavia. Four Ochagavia accessions (three species) form a polytomy in which O. andina (S81, S83) and O. litoralis (S84) share the same rDNA. On the other hand, all plastome datasets, except partitioned cp-dataset-3, placed Fascicularia within Ochagavia and O. carnea S85 within Fascicularia as a sister to F. bicolor subsp. bicolor, which is also reflected in the Neighbor-Net analyses. Conflicting topologies in plastid and nuclear phylogenetic trees could be explained by several processes such as convergent evolution, recent or historical gene flow or incomplete lineage sorting (Fehrer et al., 2007; Degnan & Rosenberg, 2009; Pelser et al., 2010). It has been shown that interspecific, as well as intergeneric, experimental hybridization in Bromelioideae is possible (e.g. Zhang et al., 2012; Souza et al., 2017), and rare natural hybridization events (Wendt et al., 2008; Matallana et al., 2016) might be important drivers of evolution in the core (Goetze et al., 2017) as well as in the early-diverging Bromelioideae (Matuszak-Renger et al., 2018). We could therefore assume that the well-resolved tree based on the slowly evolving plastomes (Smith, 2015) represents a rather gradual evolution from a common ancestor, and the position of O. carnea S85 reflects plastid capture resulting from occasional intergeneric hybridization. On the contrary, the tree based on biparentally inherited rDNA supports a rather isolated phylogenetic position of Fascicularia. This might be the result of more recent lineage sorting between these two genera, although genetic similarity among O. andina, O. carnea and O. litoralis could be explained by gene flow within Ochagavia and subsequent homogenization of the rDNA gene cluster due to concerted evolution (Kellogg & Appels, 1995; Sang, Crawford & Stuessy, 1995; Àlvarez & Wendel 2003; Calonje et al., 2009). The evolution of Fascicularia and Ochagavia has probably involved gradual evolution and several reticulation processes resulting from partially overlapping distribution ranges in the past as well as in the present (Zizka et al., 1999; Zizka et al., 2002; 2009) on the mainland c. 2.6 Mya (95% HPD 1.1–4.6) and later (Silvestro et al., 2014). Taxonomic considerations Phylogenetic methods and cladistic classification are broadly accepted in systematics, and monophyly is widely acknowledged as a criterion for grouping of taxa (De Queiroz & Gauthier, 1992). Considering Fascicularia and Ochagavia, a monophyletic group with respect to both nuclear and plastome phylogenetic trees could be achieved when including all species in one genus. In this case, the genus must carry the older name (Ochagavia; Zizka et al., 2002). However, in this case the genus would consist of taxa with contrasting morphology, which is in our opinion improper. Another possibility to achieve monophyly in concordance with phylogenetic trees based on both organelles would be the placement of the generic type O. elegans in a monotypic genus Ochagavia and the remaining species in Rhodostachys Phil. (type species R. andina Phil.), erected by Philippi in 1858. A third possibility would be to divide the sister group of O. elegans into two genera, keeping Fascicularia and combining O. andina, O. litoralis and O. carnea in Rhodostachys. The separation of three genera would be in line with altered gene content at the IRA-LSC boundary (O.elegans vs. O. carnea vs. F. bicolor), a premature stop-codon mutation very close to the 3’-end of rbcL in O. elegans and with some morphological observations. The stamens of O. elegans are not conspicuously exserted, and the species possess a well-developed epigynous tube, which reaches a length of up to 1.9 cm (Zizka et al., 2002), characters not observed elsewhere in the group. The topology of the nuclear rDNA tree seems to support the third possibility, as both Ochagavia (except O. elegans) and Fascicularia form monophyletic groups. Our Neighbor-Net rDNA analysis could be interpreted even in favour of the current taxonomic treatment as previously also suggested by mostly nuclear AFLPs (Horres et al., 2007). The rDNA analyses supported also the subspecies concept in Fascicularia, similarly as previously shown by RAPDs (Zizka et al., 1999). Biparentally inherited nuclear markers are considered more suitable for species delimitation than uniparentally inherited ones (Petit & Excoffier, 2009). However, we do not propose nomenclatural changes yet. Although there are some differences in flower morphology between O. elegans and the remaining three Ochagavia spp., these four species are similar in the majority of inflorescence and flower characters and clearly different from Fascicularia (Zizka et al., 2002). Additionally, the pollen size (c. 50 vs. c. 35 µm in the longest axis), pollen shape (oblate vs. spherical), aperture configuration (monosulcate vs. inaperturate) and the dimensions of pollen muri allow a clear separation between Ochagavia (including O. elegans) and Fascicularia (Ehler & Schill, 1973; Zöllner & Oyanedel, 1991), and are also in favour of the current taxonomic treatment (Zizka et al., 1999, 2002). Therefore, we suggest that a final decision on nomenclature and taxonomic treatment of this group should be taken when more nuclear loci and possibly also population genetic approaches are considered. Coding versus non-coding regions of plastomes and the effect of partitioning There is a clear contrast between coding and non-coding DNA regions mainly due to evolutionary constraints and natural selection, which results in a slower evolution of coding regions (e.g. Smith, 2015). Including more quickly evolving non-coding regions is particularly advantageous for phylogenetic reconstruction at a lower taxonomic level, as previously advocated by Ma et al. (2014). Indeed, in our study, much better node support and a backbone topology congruent with previous studies was achieved with the whole plastome approaches (cp-dataset-1, cp-dataset-3) than using coding genes only (cp-dataset-2). When accounting for substitution heterogeneity across different sites (partitioning), coding cp-dataset-2 revealed lower node support and altered unsupported topology. Employment of different partitioning schemes in phylogenetic reconstruction using whole plastomes was thoroughly compared by Ma et al. (2014), who showed that the tree topologies inferred from the ML analyses under various partitioning schemes remained mostly the same as in the unpartitioned analysis. However, as also observed for the partitioned cp-dataset-2, the maximally partitioned model (Partition63), which consisted of coding regions only, fits the data more poorly than the unpartitioned model. Moreover, as in our case it has been also shown that BI analysis using parameter-rich models often fail to converge and lead to long tree problem (Marshall, 2010), and varying topologies can be recovered depending on the partition analysed (Saarela et al., 2018). Our data shows that, although partitioning of plastomes provides a better fit to the data, a strong phylogenetic signal at lower taxonomic levels can robustly resolve the relationships even without the use of an optimal partitioning scheme, corroborating what was previously demonstrated in other studies (e.g. Pham et al., 2017; Dillenberger et al., 2018). CONCLUSIONS Our results provide support for monophyly of the Fascicularia-Ochagavia group and the sister position of O. elegans to the remainder of the group. Ochagavia elegans seems to have evolved in isolation since the colonization of Robinson Crusoe Island, and gene flow between the continental Fascicularia and Ochagavia occasionally occurs as suggested by the conflicting plastome and nuclear rDNA tree topologies. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher's web-site: Table S1. Collection history of studied accessions. Sample – sample number in the phylogenetic analyses, Collector, No., Date – collector, field number and date of the original collection, Source – cultivation number and IPEN if sampled in the botanical garden. Table S2.De novo plastome assembly with Fast-Plast v1.2.7 for the 15 Bromeliaceae samples. Fully assembled plastomes are shaded in grey. No reads – total number of reads, No quality reads - number of reads after quality trimming, % dropped – % of reads removed due to low quality, % aligned to plastome – % of quality reads matching to the Ananas comosus plastome, No aligned to plastome – number of plastid-derived reads, No contigs – number of resulting contigs, Longest contig length – length of the longest contig, Sum of contig lengths – total length of all reported contigs, % cpDNA genes – % of known plastid genes recovered, LSC – large single copy, SSC – small single copy, IR – inverted repeats (length in bp and average coverage is reported for these three regions), Final report – final result of the pipeline’s outcome. Figure S1. A - Best scoring ML tree based on cp-dataset-3 (full or nearly complete plastomes). Numbers at nodes show Bayesian posterior probabilities and bootstrap support percentages (PP/BS). The scale bar below the tree shows the branch length for 4.0E-4 substitutions per nucleotide position. B - Split graph resulting from Neighbor-Net analysis using uncorrected p-distances of cp-dataset-3. Numbers at the edges indicate bootstrap support. The scale bar indicates genetic distance. Figure S2. Best scoring ML tree based on cp-dataset-2 (plastome coding regions). Numbers at nodes show Bayesian posterior probabilities and bootstrap support percentages (PP/BS). The scale bar below the tree shows the branch length for 5.0E-4 substitutions per nucleotide position. Figure S3. Best scoring ML tree based on partitioned cp-dataset-2 (plastome coding regions). Numbers at nodes show Bayesian posterior probabilities and bootstrap support percentages (PP/BS). The scale bar below the tree shows the branch length for 0.02 substitutions per nucleotide position. Figure S4. Circular plots of de novo assembled plastomes with annotation. The genes shown inside and outside the outer circle are transcribed in a clockwise and counterclockwise direction, respectively. Colors denote different functional groups as shown in the legend. The hatch marks on the inner circle indicate the extent of the inverted repeats (IRA and IRB) that separate the small single copy (SSC) region from the large single copy (LSC) region. The dark gray and light gray shading within the inner circle corresponds to GC and AT content, respectively. Figure S5. Collapsed pollen in proximal polar view. A - Ochagavia andina (Zizka 8095), B - Ochagavia elegans (Zizka 1444), scale bars = 10 µm. ACKNOWLEDGEMENTS We are grateful to curators and staff of the Palmengarten Frankfurt (C. Bayer, M. Jacobi), Utrecht Botanic Gardens (E. Gouda) and Otto Zöllner (1909–2007) for providing plant material. We thank CONAF Chile for permission to collect Bromeliaceae in the natural habitat. This work was supported by the DFG [Zi 557/7-1 and 14-1]. Computational resources for the Fast-Plast analyses were provided by the CESNET LM2015042 and the CERIT Scientific Cloud LM2015085, under the programme ‘Projects of Large Research, Development, and Innovations Infrastructures’. REFERENCES Álvarez I , Wendel JF . 2003 . Ribosomal ITS sequences and plant phylogenetic inference . Molecular Phylogenetics and Evolution 29 : 417 – 434 . Google Scholar Crossref Search ADS PubMed WorldCat Baker PE , Gledhill A , Harvey PK , Hawkesworth CJ . 1987 . Geochemical evolution of the Juan Fernandez Islands, SE Pacific . Journal of the Geological Society, London 144 : 933 – 944 . Google Scholar Crossref Search ADS WorldCat Bankevich A , Nurk S , Antipov D , Gurevich AA , Dvorkin M , Kulikov AS , Lesin VM , Nikolenko SI , Pham S , Prjibelski AD , Pyshkin AV , Sirotkin AV , Vyahhi N , Tesler G , Alekseyev MA , Pevzner PA . 2012 . SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing . Journal of Computational Biology: A Journal of Computational Molecular Cell Biology 19 : 455 – 477 . Google Scholar Crossref Search ADS PubMed WorldCat Boetzer M , Henkel CV , Jansen HJ , Butler D , Pirovano W . 2011 . Scaffolding pre-assembled contigs using SSPACE . Bioinformatics (Oxford, England) 27 : 578 – 579 . Google Scholar Crossref Search ADS PubMed WorldCat Bolger AM , Lohse M , Usadel B . 2014 . Trimmomatic: a flexible trimmer for Illumina sequence data . Bioinformatics (Oxford, England) 30 : 2114 – 2120 . Google Scholar Crossref Search ADS PubMed WorldCat Bryant D , Moulton V . 2004 . Neighbor-net: an agglomerative method for the construction of phylogenetic networks . Molecular Biology and Evolution 21 : 255 – 265 . Google Scholar Crossref Search ADS PubMed WorldCat Calonje M , Martin-Bravo S , Dobeš C , Gong W , Jordon-Thaden I , Kiefer C , Kiefer M , Paule J , Schmickl R , Koch MA . 2009 . Non-coding nuclear DNA markers in phylogenetic reconstruction . Plant Systematics and Evolution 282 : 257 – 280 . Google Scholar Crossref Search ADS WorldCat Castresana J . 2000 . Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis . Molecular Biology and Evolution 17 : 540 – 552 . Google Scholar Crossref Search ADS PubMed WorldCat Degnan JH , Rosenberg NA . 2009 . Gene tree discordance, phylogenetic inference and the multispecies coalescent . Trends in Ecology & Evolution 24 : 332 – 340 . Google Scholar Crossref Search ADS PubMed WorldCat De Queiroz K , Gauthier J . 1992 . Phylogenetic taxonomy . Annual Review of Ecology, Evolution, and Systematics 23 : 449 – 480 . Google Scholar Crossref Search ADS WorldCat Dillenberger MS , Wei N , Tennessen JA , Ashman TL , Liston A . 2018 . Plastid genomes reveal recurrent formation of allopolyploid Fragaria . American Journal of Botany 105 : 862 – 874 . Google Scholar Crossref Search ADS PubMed WorldCat Doyle JJ , Doyle JL . 1990 . Isolation of plant DNA from fresh tissue . Focus 12 : 13 – 15 . OpenURL Placeholder Text WorldCat Ehler N , Schill R . 1973 . Die Pollenmorphologie der Bromeliaceae . Pollen et Spores 15 : 13 – 45 . OpenURL Placeholder Text WorldCat Evans TM , Jabaily RS , Gelli de Faria AP , Sousa LOF , Wendt T , Brown GK . 2015 . Phylogenetic relationships in bromeliaceae subfamily bromelioideae based on chloroplast DNA sequence data . Systematic Botany 40 : 116 – 128 . Google Scholar Crossref Search ADS WorldCat Fehrer J , Gemeinholzer B , Chrtek J Jr , Bräutigam S . 2007 . Incongruent plastid and nuclear DNA phylogenies reveal ancient intergeneric hybridization in Pilosella hawkweeds (Hieracium, Cichorieae, Asteraceae) . Molecular Phylogenetics and Evolution 42 : 347 – 361 . Google Scholar Crossref Search ADS PubMed WorldCat Frankham R . 1997 . Do island populations have less genetic variation than mainland populations? Heredity 78 ( Pt 3): 311 – 327 . Google Scholar Crossref Search ADS PubMed WorldCat Gelman A , Rubin D . 1992 . Inference from iterative simulation using multiple sequences . Statistical Science 7 : 457 – 472 . Google Scholar Crossref Search ADS WorldCat Gitaí J , Paule J , Zizka G , Schulte K , Benko-Iseppon AM . 2014 . Chromosome numbers and DNA content in Bromeliaceae: additional data and critical review . Botanical Journal of the Linnean Society 176 : 349 – 368 . Google Scholar Crossref Search ADS WorldCat Givnish TJ , Barfuss MHJ , Van Ee B , Riina R , Schulte K , Horres R , Gonsiska PA , Jabaily RS , Crayn DM , Smith JAC , Winter K , Brown GK , Evans TM , Holst BK , Luther H , Till W , Zizka G , Berry PE , Sytsma KJ . 2011 . Phylogeny, adaptive radiation, and historical biogeography in Bromeliaceae: insights from an eight-locus plastid phylogeny . American Journal of Botany 98 : 1 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat Goetze M , Zanella CM , Palma‐Silva C , Büttow MV , Bered F . 2017 . Incomplete lineage sorting and hybridization in the evolutionary history of closely related, endemic yellow‐flowered Aechmea species of subgenus Ortgiesia (Bromeliaceae) . American Journal of Botany 104 : 1 – 15 . Google Scholar Crossref Search ADS WorldCat Gordon A , Hannon G . 2010 . FASTX-Toolkit: FASTQ/A short-reads pre-processing tools. Available at: http://hannonlab.cshl.edu/fastx_toolkit (accessed 16 March 2018 ). Gouda EJ , Butcher D , Gouda CS . 2018. Encyclopaedia of bromeliads, Version 4 . Utrecht : University Botanic Gardens . Available at: http://bromeliad.nl/encyclopedia/ (accessed 26 February 2019 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Halbritter H . 1992 . Morphologie und systematische Bedeutung des Pollens der Bromeliaceae . Grana 31 : 197 – 212 . Google Scholar Crossref Search ADS WorldCat Halbritter H . 1998 . Preparing living pollen material for scanning electron microscopy using 2,2-dimethoxypropane (DMP) and critical-point drying . Biotechnic & Histochemistry: Official Publication of the Biological Stain Commission 73 : 137 – 143 . Google Scholar Crossref Search ADS PubMed WorldCat Halbritter H , Ulrich S , Grimsson F , Weber M , Zetter R , Hesse M , Buchner R , Svojtka M , Frosch-Radivo A . 2018 . Illustrated pollen terminology . 2nd edn. Berlin : Springer . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Horres R , Schulte K , Weising K , Zizka , G . 2007 . Systematics of Bromelioideae (Bromeliaceae) – evidence from molecular and antomical studies . Aliso 23 : 27 – 43 . Google Scholar Crossref Search ADS WorldCat Horres R , Zizka G , Kahl G , Weising K . 2000 . Molecular phylogenetics of Bromeliaceae: evidence from trnL (UAA) intron sequences of the chloroplast genome . Plant Biology 2 : 306 – 315 . Google Scholar Crossref Search ADS WorldCat Huson DH , Bryant D . 2006 . Application of phylogenetic networks in evolutionary studies . Molecular Biology and Evolution 23 : 254 – 267 . Google Scholar Crossref Search ADS PubMed WorldCat Katoh K , Standley DM . 2013 . MAFFT multiple sequence alignment software version 7: improvements in performance and usability . Molecular Biology and Evolution 30 : 772 – 780 . Google Scholar Crossref Search ADS PubMed WorldCat Kearse M , Moir R , Wilson A , Stones-Havas S , Cheung M , Sturrock S , Buxton S , Cooper A , Markowitz S , Duran C , Thierer T , Ashton B , Meintjes P , Drummond A . 2012 . Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data . Bioinformatics (Oxford, England) 28 : 1647 – 1649 . Google Scholar Crossref Search ADS PubMed WorldCat Kellogg EA , Appels R . 1995 . Intraspecific and interspecific variation in 5S RNA genes are decoupled in diploid wheat relatives . Genetics 140 : 325 – 343 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Lanfear R , Calcott B , Ho SY , Guindon S . 2012 . Partitionfinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses . Molecular Biology and Evolution 29 : 1695 – 1701 . Google Scholar Crossref Search ADS PubMed WorldCat Lanfear R , Frandsen PB , Wright AM , Senfeld T , Calcott B . 2017 . PartitionFinder 2: new methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses . Molecular Biology and Evolution 34 : 772 – 773 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Langmead B , Salzberg SL . 2012 . Fast gapped-read alignment with Bowtie 2 . Nature Methods 9 : 357 – 359 . Google Scholar Crossref Search ADS PubMed WorldCat Leme EMC , Heller S , Zizka G , Halbritter H . 2017 . New circumscription of Cryptanthus and new cryptanthoid genera and subgenera (Bromeliaceae: Bromelioideae) based on neglected morphological traits and molecular phylogeny . Phytotaxa 318 : 1 – 88 . Google Scholar Crossref Search ADS WorldCat Lohse M , Drechsel O , Bock R . 2007 . OrganellarGenomeDRAW (OGDRAW): a tool for the easy generation of high-quality custom graphical maps of plastid and mitochondrial genomes . Current Genetics 52 : 267 – 274 . Google Scholar Crossref Search ADS PubMed WorldCat Ma PF , Zhang YX , Zeng CX , Guo ZH , Li DZ . 2014 . Chloroplast phylogenomic analyses resolve deep-level relationships of an intractable bamboo tribe Arundinarieae (poaceae) . Systematic Biology 63 : 933 – 950 . Google Scholar Crossref Search ADS PubMed WorldCat Marçais G , Kingsford C . 2011 . A fast, lock-free approach for efficient parallel counting of occurrences of k-mers . Bioinformatics (Oxford, England) 27 : 764 – 770 . Google Scholar Crossref Search ADS PubMed WorldCat Marshall DC . 2010 . Cryptic failure of partitioned Bayesian phylogenetic analyses: lost in the land of long trees . Systematic Biology 59 : 108 – 117 . Google Scholar Crossref Search ADS PubMed WorldCat Matallana G , Oliveira PE , Silva PRR , Wendt T . 2016 . Post-pollination barriers in an assemblage of Bromeliaceae in south-eastern Brazil . Botanical Journal of the Linnean Society 181 : 521 – 531 . Google Scholar Crossref Search ADS WorldCat Matuszak-Renger S , Paule J , Heller S , Leme EMC , Steinbeisser GM , Barfuss MHJ , Zizka G . 2018 . Phylogenetic relationships among Ananas and related taxa (Bromelioideae, Bromeliaceae) based on nuclear, plastid and AFLP data . Plant Systematics and Evolution 304 : 841 – 851 . Google Scholar Crossref Search ADS WorldCat McKain MR , Wilson MC . 2018 . Fast-Plast: a rapid de novo assembly pipeline for whole chloroplast genomes. Available at: https://github.com/mrmckain/Fast-Plast (accessed 25 June 2018 ) Michaels SD , John MC , Amasino RM . 1994 . Removal of polysaccharides from plant DNA by ethanol precipitation . BioTechniques 17 : 274, 276 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Nashima K , Terakami S , Nishitani C , Kunihisa M , Shoda M , Takeuchi M , Urasaki N , Tarora K , Yamamoto T , Katayama H . 2015 . Complete chloroplast genome sequence of pineapple (Ananas comosus) . Tree Genetics & Genomes 11 : 60 . Google Scholar Crossref Search ADS WorldCat Nelson EC , Zizka G . 1997 . Fascicularia (Bromeliaceae): which species are cultivated and naturalized in northwestern Europe? New Plantsman 4 : 232 – 239 . OpenURL Placeholder Text WorldCat Ovcharenko I , Loots GG , Giardine BM , Hou M , Ma J , Hardison RC , Stubbs L , Miller W . 2005 . Mulan: multiple-sequence local alignment and visualization for studying function and evolution . Genome Research 15 : 184 – 194 . Google Scholar Crossref Search ADS PubMed WorldCat Pelser PB , Kennedy AH , Tepe EJ , Shidler JB , Nordenstam B , Kadereit JW , Watson LE . 2010 . Patterns and causes of incongruence between plastid and nuclear Senecioneae (Asteraceae) phylogenies . American Journal of Botany 97 : 856 – 873 . Google Scholar Crossref Search ADS PubMed WorldCat Petit RJ , Excoffier L . 2009 . Gene flow and species delimitation . Trends in Ecology & Evolution 24 : 386 – 393 . Google Scholar Crossref Search ADS PubMed WorldCat Pham KK , Hipp AL , Manos PS , Cronn RC . 2017 . A time and a place for everything: phylogenetic history and geography as joint predictors of oak plastome phylogeny . Genome 60 : 720 – 732 . Google Scholar Crossref Search ADS PubMed WorldCat Rambaut A , Drummond AJ , Xie D , Baele G , Suchard MA . 2018 . Posterior summarization in Bayesian phylogenetics using tracer 1.7 . Systematic Biology 67 : 901 – 904 . Google Scholar Crossref Search ADS PubMed WorldCat Ronquist F , Huelsenbeck JP . 2003 . MrBayes 3: Bayesian phylogenetic inference under mixed models . Bioinformatics (Oxford, England) 19 : 1572 – 1574 . Google Scholar Crossref Search ADS PubMed WorldCat Saarela JM , Burke SV , Wysocki WP , Barrett MD , Clark LG , Craine JM , Peterson PM , Soreng RJ , Vorontsova MS , Duvall MR . 2018 . A 250 plastome phylogeny of the grass family (Poaceae): topological support under different data partitions . PeerJ 6 : e4299 . Google Scholar Crossref Search ADS PubMed WorldCat Sang T , Crawford DJ , Stuessy TF . 1995 . Documentation of reticulate evolution in peonies (Paeonia) using internal transcribed spacer sequences of nuclear ribosomal DNA: implications for biogeography and concerted evolution . Proceedings of the National Academy of Sciences of the United States of America 92 : 6813 – 6817 . Google Scholar Crossref Search ADS PubMed WorldCat Schulte K , Barfuss MH , Zizka G . 2009 . Phylogeny of Bromelioideae (Bromeliaceae) inferred from nuclear and plastid DNA loci reveals the evolution of the tank habit within the subfamily . Molecular Phylogenetics and Evolution 51 : 327 – 339 . Google Scholar Crossref Search ADS PubMed WorldCat Schulte K , Horres R , Zizka G . 2005 . Molecular phylogeny of Bromelioideae and its implications on biogeography and the evolution of CAM in the family (Poales, Bromeliaceae) . Senckenbergiana Biologica 85 : 113 – 125 . OpenURL Placeholder Text WorldCat Schulte K , Zizka G . 2008 . Multi locus plastid phylogeny of Bromelioideae (Bromeliaceae) and the taxonomic utility of petal appendages and pollen characters . Candollea 63 : 209 – 225 . OpenURL Placeholder Text WorldCat Silvestro D , Michalak I . 2012 . raxmlGUI: a graphical front-end for RAxML . Organisms Diversity & Evolution 12 : 335 – 337 . Google Scholar Crossref Search ADS WorldCat Silvestro D , Zizka G , Schulte K . 2014 . Disentangling the effects of key innovations on the diversification of Bromelioideae (Bromeliaceae) . Evolution; International Journal of Organic Evolution 68 : 163 – 175 . Google Scholar Crossref Search ADS PubMed WorldCat Smith DR . 2015 . Mutation rates in plastid genomes: they are lower than you might think . Genome Biology and Evolution 7 : 1227 – 1234 . Google Scholar Crossref Search ADS PubMed WorldCat Souza EH , Versieux LM , Souza FVD , Rossi ML , Costa MAPC , Martinelli AP . 2017 . Interspecific and intergeneric hybridization in Bromeliaceae and their relationships to breeding systems . Scientia Horticulturae 223 : 53 – 61 . Google Scholar Crossref Search ADS WorldCat Stamatakis A . 2014 . RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies . Bioinformatics (Oxford, England) 30 : 1312 – 1313 . Google Scholar Crossref Search ADS PubMed WorldCat Stuessy TF , Foland KA , Sutter JF , Sanders RW , O MS . 1984 . Botanical and geological significance of potassium-argon dates from the Juan Fernandez islands . Science (New York, N.Y.) 225 : 49 – 51 . Google Scholar Crossref Search ADS PubMed WorldCat Stuessy TF , Takayama K , López-Sepúlveda P , Crawford DJ . 2014 . Interpretation of patterns of genetic variation in endemic plant species of oceanic islands . Botanical Journal of the Linnean Society. Linnean Society of London 174 : 276 – 288 . Google Scholar Crossref Search ADS PubMed WorldCat Tillich M , Lehwark P , Pellizzer T , Ulbricht-Jones ES , Fischer A , Bock R , Greiner S . 2017 . GeSeq – versatile and accurate annotation of organelle genomes . Nucleic Acids Research 45 : W6 – W11 . Google Scholar Crossref Search ADS PubMed WorldCat Wendt T , Coser TS , Matallana G , Guilherme FAG . 2008 . An apparent lack of prezygotic reproductive isolation among 42 sympatric species of Bromeliaceae in southeastern Brazil . Plant Systematics and Evolution 275 : 31 – 41 . Google Scholar Crossref Search ADS WorldCat Wilkin P . 1996 . Ochagavia elegans. Curtis’s Botanical Magazine 13 : 22 – 26 . Google Scholar Crossref Search ADS WorldCat Xu B , Zeng XM , Gao XF , Jin DP , Zhang LB . 2017 . ITS non-concerted evolution and rampant hybridization in the legume genus Lespedeza (Fabaceae) . Scientific Reports 7 : 40057 . Google Scholar Crossref Search ADS PubMed WorldCat van de Zande L , Bijlsma R . 1995 . Limitations of the RAPD technique in phylogeny reconstruction in Drosophila . Journal of Evolutionary Biology 8 : 645 – 656 . Google Scholar Crossref Search ADS WorldCat Zhang F , Ge YY , Wang WY , Shen XL , Yu XY . 2012 . Assessing genetic divergence in interspecific hybrids of Aechmea gomosepala and A. recurvata var. recurvata using inflorescence characteristics and sequence-related amplified polymorphism markers . Genetics and Molecular Research: GMR 11 : 4169 – 4178 . Google Scholar Crossref Search ADS PubMed WorldCat Zizka G , Horres R , Nelson EC , Weising K . 1999 . Revision of the genus Fascicularia Mez (Bromeliaceae) . Botanical Journal of the Linnean Society 129 : 315 – 332 . OpenURL Placeholder Text WorldCat Zizka G , Schmidt M , Schulte K , Novoa P , Pinto R , König K . 2009 . Chilean Bromeliaceae: diversity, distribution and evaluation of conservation status . Biodiversity and Conservation 18 : 2449 – 2471 . Google Scholar Crossref Search ADS WorldCat Zizka G , Trumpler K , Zöllner O . 2002 . Revision of the genus Ochagavia (Bromeliaceae, Bromelioideae) . Willdenowia 32 : 331 – 350 . Google Scholar Crossref Search ADS WorldCat Zöllner O , Oyanedel E . 1991 . A Chilean bromeliad genus of the temperate zone . Journal of the Bromeliad Society 41 : 169 – 171 . OpenURL Placeholder Text WorldCat © 2019 The Linnean Society of London, Botanical Journal of the Linnean Society This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Phylogenomic insights into the Fascicularia-Ochagavia group (Bromelioideae, Bromeliaceae)
JO - Botanical Journal of the Linnean Society
DO - 10.1093/botlinnean/boz085
DA - 2020-03-27
UR - https://www.deepdyve.com/lp/oxford-university-press/phylogenomic-insights-into-the-fascicularia-ochagavia-group-rbFWAzTzE6
DP - DeepDyve
ER -