Evolutionary relationships in Santalales inferred using target capture with Angiosperms353, focusing on Australasian Santalaceae sensu lato

Abstract

The angiosperm order Santalales comprises more than 2500 species, most of which are hemi- or holoparasitic on other plants, and derive water and nutrients via specialised structures that attach to host roots or stems. The parasitic lifestyle has affected the morphology and genomes of these plants, and classification of the order has been difficult, with outstanding questions about membership of and relationships between families in the order. We chose to focus on improving phylogenetic sampling in the broadly circumscribed Santalaceae sens. lat., with emphasis on Australasian members of Amphorogynaceae and Viscaceae as part of the Genomics for Australian Plants Initiative. We used target capture with the Angiosperms353 bait set to generate a dataset of 318 nuclear loci × 195 samples, including publicly available data from other Santalales families. Phylogenetic inferences using maximum likelihood concatenation and a summary coalescent approach were largely congruent and resolved relationships between most families, agreeing with much of the previous work on the order. Some relationships that have been difficult to resolve remained so, such as branching order among some families in Olacaceae sens. lat. and Santalaceae sens. lat. Denser sampling in Amphorogynaceae and Viscaceae provided new insights into species-level relationships in genera such as Leptomeria and Choretrum, and allowed testing of recent phylogenetic work in Korthalsella. Our new phylogenetic hypothesis is consistent with one origin of root hemiparasitism, two origins of holoparasitism and five origins of aerial parasitism in the order. Although Angiosperms353 was successful, some phylogenetic bias in gene recovery suggests that future studies may benefit from more specific baits and deeper sequencing, especially for Viscaceae.