RADseq vs. UCEs, round 3

Though reduced-representation genome sequencing (or high-throughput, or nextgen, or massively parallel sequencing, or…) has become standard practice for molecular ecology labs over the past few years, the relative merits of different library preparation methods remains an active area of research. Two of the most popular options use either restriction digests (e.g., ddRADseq, RADseq, GBS) or sequence capture with RNA probes (UCEs, exon-capture) to produce large numbers of loci that are then indexed for sequencing on Illumina platforms. These techniques target different regions of genome, vary in cost and reproducibility, and have their own vocal and partisan adherents.

Your garden-variety Piranga sp. -- the Western Tanager (Wikimedia Commons: Kati Fleming)

Your garden-variety Piranga sp. — the Western Tanager (Wikimedia Commons: Kati Fleming)

As a result, there’s been a new genre of paper in Systematic Biology and its ilk lately, ephemeral but representative of its time: the empirical phylogenetic study framed as a comparison of the efficacy of these methods. Smith et al (2013) kicked things off by exploring the performance of RADseq and UCEs at shallow timescales in a comparative phylogeographic study of Amazonian forest birds (both are decent!). Leaché et al (2015) followed suit with a study of deeper evolutionary relationships in Phyrnosomatid spiny lizards (RADseq is troublingly sensitive to bioninformatics parameters at the filtering stage!).

Most recently, JD Manthey and colleagues shoot for the middle, comparing methods in a genus of songbird (Piranga) with a not-too-shallow-or-too-deep range of divergences they see as typical in many phylogenetic studies: 0.5 – 6 million years. Using both RADseq and UCE sequence data to estimate relationships among a common set of 28 samples representing all 11 recognized Piranga species, Manthey et al. asked how trees from each differed in their phylogenetic signal and topology. For good measure, they investigated the influence of including uninformative UCE loci in their analysis as well.

Encouragingly, their results shouldn’t scare you off from whatever you were planning on doing. Relationships within Piranga as inferred from RADseq and UCE data were identical, and showed nearly identical levels of phylogenetic signal (each identified a single node with less than full support). However, be careful when including UCE loci with fewer than 10 parsimony informative sites: the bevy of different species tree methods they applied to Piranga produced contrasting strongly supported topologies.

Figure 3 from Manthey et al. (2016).

Figure 3 from Manthey et al. (2016).

In a niche of biology known for its complexity and jargon, it’s a straightforward and useful takeaway. As broad questions about performance of these techniques at different time scales begin to get resolved, the obvious next step is to explore specific biases in tree parameters other than topology in greater depth. I suspect we’ll continue to see comparisons for years to come.

References:

Smith, B.T., Harvey, M.G., Faircloth, B.C., Glenn, T.C. and Brumfield, R.T., 2013. Target capture and massively parallel sequencing of ultraconserved elements (UCEs) for comparative studies at shallow evolutionary time scales. Systematic Biology. DOI: 10.1093/sysbio/syt061

Leaché, A.D., Chavez, A.S., Jones, L.N., Grummer, J.A., Gottscho, A.D. and Linkem, C.W., 2015. Phylogenomics of Phrynosomatid lizards: conflicting signals from sequence capture versus restriction site associated DNA sequencing. Genome Biology and Evolution. DOI: 10.1093/gbe/evv026

Manthey, J.D., Campillo, L.C., Burns, K.J. and Moyle, R.G., 2016. Comparison of Target-capture and Restriction-site Associated DNA Sequencing for Phylogenomics: a Test in Cardinalid Tanagers (Aves, Genus: Piranga). Systematic Biology. DOI: 10.1093/sysbio/syw005

Share

About Ethan Linck

I'm a Ph.D. Candidate at the Department of Biology and the Burke Museum of Natural History, University of Washington, Seattle. I'm interested in population genetics, speciation, and natural history, mostly in birds.
This entry was posted in bioinformatics, evolution, genomics, methods, next generation sequencing, phylogenetics and tagged , , , . Bookmark the permalink.