RT Journal Article
SR Electronic
T1 Anchored Phylogenomics of Angiosperms I: Assessing the Robustness of Phylogenetic Estimates
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 086298
DO 10.1101/086298
A1 Chris Buddenhagen
A1 Alan R. Lemmon
A1 Emily Moriartya Lemmon
A1 Jeremy Bruhl
A1 Jennifer Cappa
A1 Wendy L. Clement
A1 Michael J. Donoghue
A1 Erika J. Edwards
A1 Andrew L. Hipp
A1 Michelle Kortyna
A1 Nora Mitchell
A1 Abigail Moore
A1 Christina J. Prychid
A1 Maria C. Segovia-Salcedo
A1 Mark P. Simmons
A1 Pamela S. Soltis
A1 Stefan Wanke
A1 Austin Mast
YR 2016
UL http://biorxiv.org/content/early/2016/11/28/086298.abstract
AB An important goal of the angiosperm systematics community has been to develop a shared approach to molecular data collection, such that phylogenomic data sets from different focal clades can be combined for meta-studies across the entire group. Although significant progress has been made through efforts such as DNA barcoding, transcriptome sequencing, and whole-plastid sequencing, the community current lacks a cost efficient methodology for collecting nuclear phylogenomic data across all angiosperms. Here, we leverage genomic resources from 43 angiosperm species to develop enrichment probes useful for collecting ~500 loci from non-model taxa across the diversity of angiosperms. By taking an anchored phylogenomics approach, in which probes are designed to represent sequence diversity across the group, we are able to efficiently target loci with sufficient phylogenetic signal to resolve deep, intermediate, and shallow angiosperm relationships. After demonstrating the utility of this resource, we present a method that generates a heat map for each node on a phylogeny that reveals the sensitivity of support for the node across analysis conditions, as well as different locus, site, and taxon schemes. Focusing on the effect of locus and site sampling, we use this approach to statistically evaluate relative support for the alternative relationships among eudicots, monocots, and magnoliids. Although the results from supermatrix and coalescent analyses are largely consistent across the tree, we find support for this deep relationship to be more sensitive to the particular choice of sites and loci when a supermatrix approach as employed. Averaged across analysis approaches and data subsampling schemes, our data support a eudicot-monocot sister relationship, which is supported by a number of recent angiosperm studies.