Supplementary material for: Phylogenomics and historical biogeography of seahorses, dragonets, goatfishes, and allies (Teleostei: Syngnatharia): Assessing factors driving uncertainty in biogeographic inferences

The charismatic trumpetfishes, goatfishes, dragonets, flying gurnards,
seahorses, and pipefishes encompass a recently defined yet extraordinarily
diverse clade of percomorph fishes—the series Syngnatharia. This group is
widely distributed in tropical and warm-temperate regions, with a great
proportion of its extant diversity occurring in the Indo-Pacific. Because
most syngnatharians feature long-range dispersal capabilities, tracing
their biogeographic origins is challenging. Here, we applied an
integrative phylogenomic approach to elucidate the evolutionary
biogeography of syngnatharians. We built upon a recently published
phylogenomic study that examined ultraconserved elements by adding 62
species (total 169 species) and one family (Draconettidae), to cover ca.
25% of the species diversity and all 10 families in the group. We inferred
a set of time-calibrated trees and conducted ancestral range estimations.
We also examined the sensitivity of these analyses to phylogenetic
uncertainty (estimated from multiple genomic subsets), area delimitation,
and biogeographic models that include or exclude the jump-dispersal
parameter (j). Of the three factors examined, we found that the j
parameter has the strongest effect in ancestral range estimates, followed
by number of areas defined, and tree topology and divergence times. After
accounting for these uncertainties, our results reveal that syngnatharians
originated in the ancient Tethys Sea ca. 87 Ma (84­–94 Ma; Late
Cretaceous) and subsequently occupied the Indo-Pacific. Throughout
syngnatharian history, multiple independent lineages colonized the eastern
Pacific (6–8 times) and the Atlantic (6–14 times) from their center of
origin, with most events taking place following an east-to-west route
prior to the closure of the Tethys Seaway ca. 12–18 Ma. Ultimately, our
study highlights the importance of accounting for different factors
generating uncertainty in macroevolutionary and biogeographic inferences.