Anchored hybrid enrichment generated nuclear, plastid and mitochondrial markers resolve the Lepanthes horrida (Orchidaceae: Pleurothallidinae) species complex

Abstract

Phylogenetic relationships in species complexes and lineages derived from rapid diversifications are often challenging to resolve using morphology or standard DNA barcoding markers. The hyper-diverse genus Lepanthes from Neotropical cloud forest includes over 1200 species and many recent, explosive diversifications that have resulted in poorly supported nodes and morphological convergence across clades. Here, we assess the performance of 446 nuclear-plastid-mitochondrial markers derived from an anchored hybrid enrichment approach (AHE) coupled with coalescence- and species network-based inferences to resolve phylogenetic relationships and improve species recognition in the Lepanthes horrida species group. In addition to using orchid-specific probes to increase enrichment efficiency, we improved gene tree resolution by extending standard angiosperm targets into adjacent exons. We found high topological discordance among individual gene trees, suggesting that hybridization/polyploidy may have promoted speciation in the lineage via formation of new hybrid taxa. In addition, we identified ten loci with the highest phylogenetic informativeness values from these genomes. Most previous phylogenetic sampling in the Pleurothallidinae relies on two regions (ITS and matK), therefore, the evaluation of other markers such as those shown here may be useful in future phylogenetic studies in the orchid family. Coalescent-based species tree estimation methods resolved the phylogenetic relationships of the L. horrida species group. The resolution of the phylogenetic estimations was improved with the inclusion of extended anchor targets. This approach produced longer loci with higher discriminative power. These analyses also disclosed two undescribed species, L. amicitiae and L. genetoapophantica, formally described here, which are also supported by morphology. Our study demonstrates the utility of combined genomic evidence to disentangle phylogenetic relationships at very shallow levels of the tree of life, and in clades showing convergent trait evolution. With a fully resolved phylogeny, is it possible to disentangle traits evolving in parallel or convergently across these orchid lineages such as flower color and size from diagnostic traits such as the shape and orientation of the lobes of the petals and lip.

An empirical assessment of a single family‐wide hybrid capture locus set at multiple evolutionary timescales in Asteraceae

“…Alternatively, the following approaches do not require genomic location information: guenomu, a Bayesian hierarchical model that estimates species trees from unrooted gene trees from multiple gene families (de Oliveira Martins and Posada, ), and a recent model within PhyloNet that incorporates incomplete lineage sorting and gene duplication and loss (Du et al., ). Conflict analyses allow further investigation into discordance between gene and species trees and detection of outlier gene trees for large genomic data sets, for example, using the software phyparts (Smith et al., ), which has been used for conflict analyses in a number of lineages, including Pleurothallidinae (Orchidaceae; Bogarín et al., ), Portulacineae (Wang et al., ), Caryophyllales (Walker et al., ), and Metazoa (Shen et al., ). At lower taxonomic levels, network approaches might supersede tree‐based approaches due to the large extent of reticulation in such data sets.…”

Resolving relationships in an exceedingly young orchid lineage using Genotyping-by-sequencing data

“…In the case of orchids, AHE did not yield loci alignments with PI values equal to 0. Therefore, they were more informative than the GBS loci sequenced here (Bogarín et al, 2018). Moreover, the AHE loci resolved intricate phylogenetic relationships with confidence in a taxonomically difficult complex of a recent linage of Lepanthes orchids.…”

“…Lack of phylogenetic signal in the loci data, or rapid diversification events, arefurther thought to produce low statistical support (Bogarín et al, 2018). Low PI profiles are derived from sites evolving at rates that are either extremely fast or extremely slow (Townsend 2007).…”

Phylogenomics within the Anthonotha clade (Detarioideae, Leguminosae) reveals a high diversity in floral trait shifts and a general trend towards organ number reduction

“…Target enrichment is a cost-effective strategy to obtain markers for phylogenetic analysis at multiple taxonomic levels, including a set of markers for the entire flowering plants (angiosperms) (Budenhagen et al, 2016; Johnson et al, 2018). This approach has been successfully applied in several plant families (Carlsen et al, 2018; Comer et al, 2016; Herrando-Morairaa et al, 2018; Mandel et al, 2015, 2014; Moore et al, 2018) and it has proved useful to reconstruct relationships among closely related genera, within the same genus (Bogarin et al, 2018; Fragoso-Martíneza et al, 2017; Heyduk et al, 2016; Mitchell et al, 2017; Schmickl et al, 2016) and at species level (Nicholls et al, 2015; Villaverde et al, 2018). Within Leguminosae, this approach has been applied on lineages of the Papilionoideae and Caesalpinoideae (De Sousa et al, 2014; Nicholls et al, 2015; Ogutcen et al, 2018; Vatanparast et al, 2018) and our study is the first to use target enrichment outside these subfamilies using a complete sampling representing the entire extant diversity of the target group.…”