Patterns of clade support across the major lineages of moss phylogeny

Abstract

Relationships among the major branches of moss phylogeny are understudied compared with other major land-plant groups. We addressed this by surveying 14-17 plastid genes from taxa representing the major lineages, using different phylogenetic methods (parsimony, likelihood) and codon-and gene-based data partitioning schemes (likelihood). Our phylogenetic inferences generally corroborated the best supported clades across multiple recent studies, with comparable or higher levels of clade support here. We resolved persistent ambiguities with strong to moderate support across analyses, including several early nodes in subclass Dicranidae, and relationships among other subclasses of peristomate mosses. In particular, we resolved a sister-group relationship between Bryidae and Dicranidae, between these subclasses and Timiidae, and between this entire clade and Funariidae. We consistently recovered Tetraphidopsida (a nematodontous class) as the sister group of arthrodontous mosses (Bryopsida), although with only weak support. Strongly conflicting arrangements at the base of moss phylogeny concerning Takakiopsida and Sphagnopsida, two non-peristomate moss lineages, were inferred in parsimony and likelihood analysis, but this depended on how base-frequency parameters were estimated and how data were partitioned in likelihood analysis. Relationships inferred for the remaining peristomate and non-peristomate moss clades, and their associated support values, were otherwise broadly congruent across analyses.

Wanted dead or alive (probably dead): Stem group Polytrichaceae

“…In summary, our experiments with outgroup sampling regimes confirm the results of previous morphological (Smith, 1971;Smith Merrill, 2007) and molecular phylogenetic studies (Bell and Hyvönen, 2008;Chang and Graham, 2014), which indicate that polytrichaceous mosses represent a morphologically unique group with no known close living relatives (long branch). Our experiments also suggest that there is no way to root phylogenetic analyses of the Polytrichaceae that does not distort ingroup relationships.…”

“…3). Whereas this placement of Oedipodium is congruent with the results of recent molecular phylogenetic studies (Chang and Graham, 2014), Tetraphis is not recovered as sister to Oedipodium in these studies. The ingroup root is between a clade consisting of Psilopilum and Notoligotrichum and the rest of Polytrichaceae (Figs.…”

“…In contrast, molecular phylogenetic studies have yielded hypotheses of relationships within the family with high support values for most nodes (e.g., Bell and Hyvönen, 2010a). Nevertheless, phylogenetic relationships of the Polytrichaceae with other acrocarpous moss lineages, as inferred from molecular data are more equivocal (Chang and Graham, 2014). Because the family represents the tip of a very long branch with no close living relatives, rooting phylogenetic analyses of the Polytrichaceae (both molecular and morphological) has proved difficult and several rooting strategies have been employed in different studies.…”

Organellomic data sets confirm a cryptic consensus on (unrooted) land‐plant relationships and provide new insights into bryophyte molecular evolution

“…(, , ) demonstrated that data from even highly reduced and rapidly evolving plastid genes or plastomes can be used to place heterotrophic plants with confidence. We were particularly interested in characterizing the Buxbaumia plastome in light of its purported heterotrophic status, given the placement of this small genus as sister to all other members of Bryopsida, which is by far the most species‐rich class of mosses (e.g., Qiu et al., , ; Chang and Graham, , ; Liu et al., , ; Gitzendanner et al., ).…”

“…In the mosses, molecular analyses have also resolved many of the deep relationships, but there are still several critical nodes that remain contentious (e.g., Newton et al., ; Cox et al., ; Qiu et al., ; Chang and Graham, , ; Liu et al., ). Of these, the relationships among Takakiopsida, Sphagnopsida, and the rest of the mosses (i.e., whether Takakiopsida is sister to all other mosses or forms a clade with Sphagnopsida) are perhaps the most intriguing, as conflicting topologies are often recovered with strong support (Qiu et al., ; Chang and Graham, , ; Liu et al., , ; One Thousand Plant Transcriptomes Initiative, ). The source of this conflict is uncertain, but Chang and Graham () observed that relationships were sensitive to different model or method assumptions.…”