Data from: Biogeographic analysis reveals ancient continental vicariance and recent oceanic dispersal in amphibians

Amphibians comprise over 7000 extant species distributed in almost every ecosystem on every continent except Antarctica. Most species also show high specificity for particular habitats, biomes, or climatic niches, seemingly rendering long-distance dispersal unlikely. Indeed, many lineages still seem to show the signature of their Pangaean origin, ~300Ma later. To date, no study has attempted a large-scale historical-biogeographic analysis of the group to understand the distribution of extant lineages. Here, I use an updated chronogram containing 3309 species (~45% of extant diversity) to reconstruct their movement between 12 global ecoregions. I find that a Pangaean origin and subsequent Laurasian and Gondwanan fragmentation explains a large proportion of patterns in the distribution of extant species. However, dispersal during the Cenozoic, likely across land bridges or short distances across oceans, has also exerted a strong influence. Finally, there are at least three strongly supported instances of long-distance oceanic dispersal between former Gondwanan landmasses during the Cenozoic. Intermediate extinction from intervening areas seems to be a strong factor in shaping present-day distributions. Both dispersal and intermediate extinction are apparently tied to the evolution of extraordinarily adaptive expansion-oriented phenotypes (allowing lineages to easily colonize new areas and speciate), or conversely, to extremely specialized phenotypes or heavily relictual climatic niches that result in strong geographic localization and limited diversification.Amphibia_2_9Matrix of 3309 amphibian species sampled for up to 12 genes (12809bp total) and one outgroup.Amphibia_ModelsModel file for the Amphibia_2_9 matrix, describing the 12 genes and their partitions.amph_shlNNI-optimized ML tree resulting from Amphibia_2_9 matrix, with SHL support values.amph_shl_datesDated version of the NNI-optimized ML tree with SHL support values, used for biogeographic reconstructions.amph_clade_areasData illustrated in Figure 1: clade names, clade ages, number of areas occupied, and number of species in clade.amph_ecoregionsTaxonomic database listing 6576 species, their family, and regions of occurrence.amph_shl_dates_trim.treMESQUITE-format NEXUS file containing the tree pruned to the 3126 species inhabiting one ecoregion, with Mk1 estimates of ancestral states (Appendix S5).caecilian_ecoregionsEcoregion data for BioGeoBEARS analysis of caecilian subtree.caecilian_ecoregionsR Code for BioGeoBEARS run on caecilian subtree (Figure 1).amph_shl_dates_caeciliansCaecilian subtree for BioGeoBEARS analysis (Figure 1).caecilian_DEC_M0_unconstrained_v1Output for DEC model from caecilian subtree analysis.caecilian_DECJ_M0_unconstrained_v1Output for DECj model from caecilian subtree analysis.salamander_ecoregionsSalamander ecoregion data for 6-area BioGeoBEARS run. Code given below.salamander_ecoregionsR Code for BioGeoBEARS analysis of salamander subtree with 6 areas. Data produced are given below.amph_shl_dates_salamandersSalamander subtree from NNI-optimized, SHL-supported ML tree. Used for 6-area salamander analysis in BioGeoBEARS.salamander_DEC_M0_unconstrained_v1R data object produced by R code given above, from 6-area salamander-subtree analysis in BioGeoBEARS. Load to view ancestral area estimates.salamander_DECJ_M0_unconstrained_v1R data object from DECj analysis of 6-area salamander subtree in BioGeoBEARS. Load for ancestral-area estimates.frog_ecoregions_reducedEcoregion data for 2875-taxon subtree analysis of frogs (Appendix S3).frog_ecoregions_reducedR Code for frog subtree analysis using BioGeoBEARS (Appendix S3).amph_shl_dates_frogs2875-taxon subtree for frogs (Appendix S3).frog_DEC_M0_unconstrained_v1R Data object containing results from DEC analysis of frog subtree in BioGeoBEARS (Appendix S3). Load this for ancestral-state estimates (see R code above).frog_DECJ_M0_unconstrained_v1R Data object from DECj model of frog subtree. Load for ancestral-state estimates.amph_ecoregions_reducedEcoregion data for 3309-taxon whole-tree analysis of amphibians in BioGeoBEARS (Appendix S4).amph_ecoregions_reducedR code for whole-tree analysis of amphibians (Appendix S4) in BioGeoBEARS.amph_shl_dates_trim3309-taxon amphibian tree for whole-tree analysis using BioGeoBEARS. This tree is "amph_shl_dates.tre" trimmed of the outgroup Homo_sapiens.amph_DEC_M0_unconstrained_v1R Data object from DEC analysis of amphibians (Appendix S4).amph_DECJ_M0_unconstrained_v1R Data object for DECj analysis of amphibians as a whole (Appendix S4). Load for ancestral-state estimates.New_NamesThe taxonomic database used here is from the 2009 update of AmphibiaWeb (http://www.amphibiaweb.org/). The genus-level taxonomy of amphibians is in constant flux recently, and many genera have changed in the last few years. This document translates the taxon labels in the 3309-taxon matrix into the name appearing in a 2014 update of AmphibiaWeb. Notably, the genera _Rana_ and _Bufo_ were broken up by Frost et al. (2006) and others, and the monophyly of these groups is actually supported in the current tree using their names. Some other genera are still not monophyletic, even using recently updated names. However, for those wishing to use the trees presented here for comparative analyses, this chart shows the most recent "correct" names that should be used, and for which data may have been gathered.READMEGuide to the supplemental materials.Pyron_Appendix_S1Appendix S1: caecilian subtree (DEC/DECj).Pyron_Appendix_S2Appendix S2: salamander subtree (DEC/DECj).Pyron_Appendix_S3Appendix S3: frog subtree (DEC/DECj).Pyron_Appendix_S4Appendix S4: whole-tree analysis of amphibians (DEC/DECj).Pyron_Appendix_S5Appendix S5: whole-tree analysis of amphibians (Mk1: MESQUITE)