Department of Natural History, Royal Ontario Museum, 100 Queen's Park, Toronto, ON, Canada M5S 2C6, and Department of Ecology and Evolutionary Biology, University of Toronto, Toronto, ON, Canada M5S 3B2.
Mol Ecol Resour. 2009 May;9 Suppl s1:257-68. doi: 10.1111/j.1755-0998.2009.02650.x.
General criticisms of a single mtDNA gene barcodes include failure to identify newly evolved species, use of species-delimitation thresholds, effects of selective sweeps and chance occurrence of reciprocal monophyly within species, inability to deal with hybridization and incomplete lineage sorting, and superiority of multiple genes in species identification. We address these criticisms in birds because most species are known and thus provide an ideal test data set, and we argue with selected examples that with the exception of thresholds these criticisms are not problematic for avian taxonomy. Even closely related sister species of birds have distinctive COI barcodes, but it is not possible to universally apply distance thresholds based on ratios of within-species and among-species variation. Instead, more rigorous methods of species delimitation should be favoured using coalescent-based techniques that include tests of chance reciprocal monophyly, and times of lineage separation and sequence divergence. Incomplete lineage sorting is also easily detected with DNA barcodes, and usually at a younger time frame than a more slowly evolving nuclear gene. Where DNA barcodes detect divergent reciprocally monophyletic lineages, the COI sequences can be combined with multiple nuclear genes to distinguish between speciation or population subdivision arising from high female philopatry or regional selective sweeps. Although selective sweeps are increasingly invoked to explain patterns of shallow within-species coalescences in COI gene trees, caution is warranted in this conjecture because of limited sampling of individuals and the reduced power to detect additional mtDNA haplotypes with one gene.
通用的对单一 mtDNA 基因条码的批评包括无法识别新进化的物种、使用物种界定阈值、选择清扫和物种内互惠单系发生的机会出现、无法处理杂交和不完全谱系排序，以及多个基因在物种识别中的优越性。我们在鸟类中解决这些批评，因为大多数物种都是已知的，因此提供了一个理想的测试数据集，并且我们用选定的例子来说明，除了阈值之外，这些批评对于鸟类分类学来说并不是问题。即使是鸟类的密切相关的姐妹物种也有独特的 COI 条码，但不可能基于种内和种间变异比普遍应用距离阈值。相反，应该优先使用基于合并的技术来进行更严格的物种界定，包括对机会互惠单系发生的测试，以及谱系分离和序列分歧的时间。不完全谱系排序也很容易用 DNA 条码检测到，而且通常比进化较慢的核基因更早。在 DNA 条码检测到分歧的互惠单系谱系的地方，可以将 COI 序列与多个核基因结合使用，以区分由于雌性高度亲代选择或区域选择清扫而导致的物种形成或种群细分。尽管选择性清扫越来越多地被用来解释 COI 基因树中物种内聚合并集的浅层次模式，但由于个体的有限采样和用一个基因检测额外的 mtDNA 单倍型的能力降低，因此对这种推测应保持谨慎。
