ddRAD-seq phylogenetics based on nucleotide, indel, and presence–absence polymorphisms: Analyses of two avian genera with contrasting histories

Abstract

Genotype-by-sequencing (GBS) methods have revolutionized the field of molecular ecology, but their application in molecular phylogenetics remains somewhat limited. In addition, most phylogenetic studies based on large GBS data sets have relied on analyses of concatenated data rather than species tree methods that explicitly account for genealogical stochasticity among loci. We explored the utility of "double-digest" restriction site-associated DNA sequencing (ddRAD-seq) for phylogenetic analyses of the Lagonosticta firefinches (family Estrildidae) and the Vidua brood parasitic finches (family Viduidae). As expected, the number of homologous loci shared among samples was negatively correlated with genetic distance due to the accumulation of restriction site polymorphisms. Nonetheless, for each genus, we obtained data sets of ∼3000 loci shared in common among all samples, including a more distantly related outgroup taxon. For all samples combined, we obtained >1000 homologous loci despite ∼20my divergence between estrildid and parasitic finches. In addition to nucleotide polymorphisms, the ddRAD-seq data yielded large sets of indel and locus presence-absence polymorphisms, all of which had higher consistency indices than mtDNA sequence data in the context of concatenated parsimony analyses. Species tree methods, using individual gene trees or single nucleotide polymorphisms as input, generated results broadly consistent with analyses of concatenated data, particularly for Lagonosticta, which appears to have a well resolved, bifurcating history. Results for Vidua were also generally consistent across methods and data sets, although nodal support and results from different species tree methods were more variable. Lower gene tree congruence in Vidua is likely the result of its unique evolutionary history, which includes rapid speciation by host shift and occasional hybridization and introgression due to incomplete reproductive isolation. We conclude that ddRAD-seq is a cost-effective method for generating robust phylogenetic data sets, particularly for analyses of closely related species and genera.

Double‐digest RADseq loci using standard Illumina indexes improve deep and shallow phylogenetic resolution of Lophodermium, a widespread fungal endophyte of pine needles

“…Hence, this ddRADseq method is especially effective in revealing within‐species fungal (<97% ITS similarity) diversity, but is also cost‐effective for among‐species (e.g., 97–70% ITS similarity) diversity. However, restriction site polymorphism will likely limit the efficacy of ddRADseq for phylogenetic reconstruction beyond the genus and family levels (DaCosta & Sorenson, 2016). Furthermore, the mutation risk hypothesis suggests that fewer homologous or shared loci will be recovered for loci with greater risk of mutations.…”

“…Our case study shows that small portions of the genome sequenced with the MiSeq platform can be sufficient to produce comparative homologous loci within highly diverse taxa with small haploid genomes (Figures 6 and 7, Table S1H). Other ddRADseq datasets that are not based on strict nucleotide polymorphisms, such as indel or presence–absence polymorphisms, may further help the resolution of more distantly related lineages (DaCosta & Sorenson, 2016). …”

A tree of geese: A phylogenomic perspective on the evolutionary history of True Geese

“…Whole genome analyses have recently been used to unravel the phylogenetic relations between the major bird orders (Jarvis et al, 2014;Prum et al, 2015), but our study presents one of the first phylogenomic analyses of a group of closely related bird species (DaCosta and Sorenson, 2016;Nater et al, 2015). As both concate nation and consensus methods resulted in the same topology and all nodes on the concatenation tree are supported by high boot strap values, it is likely that the resulting phylogeny closely approaches the actual species tree for the Anserini.…”

Full Issue

“…It requires no prior genomic resources and identifies single nucleotide polymorphisms (SNPs) in close proximity to restriction enzyme cut sites (McCormack et al 2013, Toews et al 2016. The method has been increasingly applied to phylogenetic assessment, where it has shown great promise for resolving phylogenetic and phylogeographic problems, as well as species delimitation questions, in a range of taxonomic groups (Emerson et al 2010, DaCosta and Sorenson 2016, Herrera and Shank 2016. The method is best suited to groups that have diverged relatively recently (McCormack et al 2013, DaCosta and Sorenson 2016, Toews et al 2016, as the number of orthologous restriction sites decreases with increasing divergence (Herrera andShank 2016, Toews et al 2016).…”

“…The method has been increasingly applied to phylogenetic assessment, where it has shown great promise for resolving phylogenetic and phylogeographic problems, as well as species delimitation questions, in a range of taxonomic groups (Emerson et al 2010, DaCosta and Sorenson 2016, Herrera and Shank 2016. The method is best suited to groups that have diverged relatively recently (McCormack et al 2013, DaCosta and Sorenson 2016, Toews et al 2016, as the number of orthologous restriction sites decreases with increasing divergence (Herrera andShank 2016, Toews et al 2016). For these reasons, it is an appropriate method for analysis of evolutionary relationships among rosellas in the subgenus Violania, which have been proposed to have diverged since the Pleistocene, and potentially as recently as 0.1617 mya (but see Shipham et al [2015] for assumptions and limitations associated with this date).…”