Museum of Natural Science and Department of Biological Sciences, Louisiana State University, Baton Rouge, LA, USA.
Natural History Museum of Denmark, Center for Global Mountain Biodiversity, and Center for Macroecology, Evolution, and Climate, University of Copenhagen, Copenhagen, Denmark; Department of Biology, University of Florida, Gainesville, FL, USA.
Mol Phylogenet Evol. 2022 Oct;175:107559. doi: 10.1016/j.ympev.2022.107559. Epub 2022 Jul 5.
As phylogenomics focuses on comprehensive taxon sampling at the species and population/subspecies levels, incorporating genomic data from historical specimens has become increasingly common. While historical samples can fill critical gaps in our understanding of the evolutionary history of diverse groups, they also introduce additional sources of phylogenomic uncertainty, making it difficult to discern novel evolutionary relationships from artifacts caused by sample quality issues. These problems highlight the need for improved strategies to disentangle artifactual patterns from true biological signal as historical specimens become more prevalent in phylogenomic datasets. Here, we tested the limits of historical specimen-driven phylogenomics to resolve subspecies-level relationships within a highly polytypic family, the New World quails (Odontophoridae), using thousands of ultraconserved elements (UCEs). We found that relationships at and above the species-level were well-resolved and highly supported across all analyses, with the exception of discordant relationships within the two most polytypic genera which included many historical specimens. We examined the causes of discordance and found that inferring phylogenies from subsets of taxa resolved the disagreements, suggesting that analyzing subclades can help remove artifactual causes of discordance in datasets that include historical samples. At the subspecies-level, we found well-resolved geographic structure within the two most polytypic genera, including the most polytypic species in this family, Northern Bobwhites (Colinus virginianus), demonstrating that variable sites within UCEs are capable of resolving phylogenetic structure below the species level. Our results highlight the importance of complete taxonomic sampling for resolving relationships among polytypic species, often through the inclusion of historical specimens, and we propose an integrative strategy for understanding and addressing the uncertainty that historical samples sometimes introduce to phylogenetic analyses.
系统发生基因组学侧重于在物种和种群/亚种水平上进行全面的分类群采样，因此越来越多地将来自历史标本的基因组数据纳入其中。虽然历史样本可以填补我们对不同群体进化历史理解的关键空白，但它们也引入了更多的系统发生不确定性来源，使得很难从由样本质量问题引起的伪迹中辨别出新的进化关系。这些问题突出表明，随着历史标本在系统发生数据集变得更加普遍，需要改进策略来区分人为模式和真实的生物学信号。在这里，我们使用数千个超保守元件 (UCE) 测试了历史标本驱动的系统发生基因组学在解决高度多态科（新大陆鹌鹑科）亚种水平关系方面的极限。我们发现，除了包含许多历史标本的两个最多态属内的不一致关系外，种级和种级以上的关系得到了很好的解决，并且得到了高度支持。我们研究了不一致的原因，发现从分类群子集推断系统发育可以解决分歧，这表明分析亚分支可以帮助去除包含历史样本的数据集中人为原因引起的分歧。在亚种水平上，我们发现两个最多态属内具有良好分辨率的地理结构，包括该科中最多态的物种——北方鹌鹑（Colinus virginianus），这表明 UCE 内的可变位点能够解决种以下水平的系统发育结构。我们的结果强调了完整的分类群采样对于解决多态种之间关系的重要性，这种关系通常是通过包括历史标本来实现的，我们提出了一种综合策略来理解和解决历史标本有时会给系统发生分析带来的不确定性。
