Frog phylogeny: A time-calibrated, species-level tree based on hundreds of loci and 5,242 species

• Published in: Molecular phylogenetics and evolution Vol. 188; p. 107907
• Main Authors: Portik, Daniel M., Streicher, Jeffrey W., Wiens, John J.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.11.2023
• Subjects: Anura; Anurans; data collection; Divergence times; frogs; Missing data; Phylogenomics; phylogeny; species; statistical analysis; Supermatrix; trees; Missing data; Divergence times; Anurans; Supermatrix; Phylogenomics
• DOI: 10.1016/j.ympev.2023.107907
• ISSN: 1055-7903; 1095-9513

[Display omitted] •We inferred a new time-calibrated phylogeny of frogs that combines phylogenomic data with a new large-scale supermatrix.•This tree contains 5,242 species, with 2,175 species more than the previous largest supermatrix analysis.•Higher-level relationships are largely concordant with recent phylogenomic analyses, unlike recent supermatrix analyses. Large-scale, time-calibrated phylogenies from supermatrix studies have become crucial for evolutionary and ecological studies in many groups of organisms. However, in frogs (anuran amphibians), there is a serious problem with existing supermatrix estimates. Specifically, these trees are based on a limited number of loci (15 or fewer), and the higher-level relationships estimated are discordant with recent phylogenomic estimates based on much larger numbers of loci. Here, we attempted to rectify this problem by generating an expanded supermatrix and combining this with data from phylogenomic studies. To assist in aligning ribosomal sequences for this supermatrix, we developed a new program (TaxonomyAlign) to help perform taxonomy-guided alignments. The new combined matrix contained 5,242 anuran species with data from 307 markers, but with 95% missing data overall. This dataset represented a 71% increase in species sampled relative to the previous largest supermatrix analysis of anurans (adding 2,175 species). Maximum-likelihood analyses generated a tree in which higher-level relationships (and estimated clade ages) were generally concordant with those from phylogenomic analyses but were more discordant with the previous largest supermatrix analysis. We found few obvious problems arising from the extensive missing data in most species. We also generated a set of 100 time-calibrated trees for use in comparative analyses. Overall, we provide an improved estimate of anuran phylogeny based on the largest number of combined taxa and markers to date. More broadly, we demonstrate the potential to combine phylogenomic and supermatrix analyses in other groups of organisms.