Data from: Sequence capture versus restriction site associated DNA sequencing for shallow systematics

Sequence capture and restriction site associated DNA sequencing (RAD-Seq) are two genomic enrichment strategies for applying next-generation sequencing technologies to systematics studies. At shallow timescales, such as within species, RAD-Seq has been widely adopted among researchers, although there has been little discussion of the potential limitations and benefits of RAD-Seq and sequence capture. We discuss a series of issues that may impact the utility of sequence capture and RAD-Seq data for shallow systematics in non-model species. We review prior studies that used both methods, and investigate differences between the methods by re-analyzing existing RAD-Seq and sequence capture datasets from a Neotropical bird (Xenops minutus). We suggest that the strengths of RAD-Seq datasets for shallow systematics are the wide dispersion of markers across the genome, the relative ease and cost of laboratory work, the deep coverage and read overlap at recovered loci, and the high overall information that results. Sequence capture's benefits include flexibility and repeatability in the genomic regions targeted, success using low-quality samples, more straightforward read orthology assessment, and higher per-locus information content. The utility of a method in systematics, however, rests not only on its performance within a study, but on the comparability of datasets and inferences with those of prior work. In RAD-Seq datasets, comparability is compromised by low overlap of orthologous markers across species and the sensitivity of genetic diversity in a dataset to an interaction between the level of natural heterozygosity in the samples examined and the parameters used for orthology assessment. In contrast, sequence capture of conserved genomic regions permits interrogation of the same loci across divergent species, which is preferable for maintaining comparability among datasets and studies for the purpose of drawing general conclusions about the impact of historical processes across biotas. We argue that sequence capture should be given greater attention as a method of obtaining data for studies in shallow systematics and comparative phylogeography.

序列捕获（Sequence capture）与限制性位点相关DNA测序（Restriction site associated DNA sequencing, RAD-Seq）是将下一代测序（next-generation sequencing）技术应用于系统分类学研究的两种基因组富集策略。在物种内这类较浅的时间尺度下，RAD-Seq已被研究者广泛采用，但目前鲜有针对RAD-Seq与序列捕获的潜在局限与优势展开的讨论。我们探讨了一系列可能影响非模式物种浅源系统分类学研究中序列捕获和RAD-Seq数据实用性的问题。我们回顾了使用这两种方法的既往研究，并通过重新分析新热带鸟类小灶鸟（Xenops minutus）的现有RAD-Seq与序列捕获数据集，探究了两种方法间的差异。我们认为，RAD-Seq数据集用于浅源系统分类学研究的优势在于：标记在基因组中广泛分布、实验室操作相对简便且成本较低、回收位点具备较高的测序深度与读段重叠度，以及整体信息量较高。序列捕获的优势则包括：靶向基因组区域的灵活性与可重复性、对低质量样本的适配性更强、读段同源性评估更为简便，以及每位点的信息量更高。不过，某一方法在系统分类学研究中的实用性，不仅取决于其在单次研究中的表现，还取决于其数据集与推断结果能否与既往研究的数据集及推断结果相比较。在RAD-Seq数据集中，跨物种同源标记的重叠度较低，且数据集内的遗传多样性会受到所检测样本的自然杂合水平与同源性评估参数间相互作用的影响，这些因素都会削弱其可比性。与之相反，对保守基因组区域的序列捕获可实现不同物种间相同位点的靶向测序，这更便于维持不同数据集与研究间的可比性，从而能够针对不同生物群的历史过程影响得出一般性结论。我们主张，在浅源系统分类学与比较系统地理学研究中，应更多关注序列捕获这一数据获取方法。