Platt Roy N, Faircloth Brant C, Sullivan Kevin A M, Kieran Troy J, Glenn Travis C, Vandewege Michael W, Lee Thomas E, Baker Robert J, Stevens Richard D, Ray David A
Department of Biological Sciences, Texas Tech University, 2901 Main St, Lubbock, TX, USA.
Department of Biological Sciences and Museum of Natural Science, Louisiana State University, 202 Life Science Building, Baton Rouge, LA, USA.
Syst Biol. 2018 Mar 1;67(2):236-249. doi: 10.1093/sysbio/syx070.
The rapid diversification of Myotis bats into more than 100 species is one of the most extensive mammalian radiations available for study. Efforts to understand relationships within Myotis have primarily utilized mitochondrial markers and trees inferred from nuclear markers lacked resolution. Our current understanding of relationships within Myotis is therefore biased towards a set of phylogenetic markers that may not reflect the history of the nuclear genome. To resolve this, we sequenced the full mitochondrial genomes of 37 representative Myotis, primarily from the New World, in conjunction with targeted sequencing of 3648 ultraconserved elements (UCEs). We inferred the phylogeny and explored the effects of concatenation and summary phylogenetic methods, as well as combinations of markers based on informativeness or levels of missing data, on our results. Of the 294 phylogenies generated from the nuclear UCE data, all are significantly different from phylogenies inferred using mitochondrial genomes. Even within the nuclear data, quartet frequencies indicate that around half of all UCE loci conflict with the estimated species tree. Several factors can drive such conflict, including incomplete lineage sorting, introgressive hybridization, or even phylogenetic error. Despite the degree of discordance between nuclear UCE loci and the mitochondrial genome and among UCE loci themselves, the most common nuclear topology is recovered in one quarter of all analyses with strong nodal support. Based on these results, we re-examine the evolutionary history of Myotis to better understand the phenomena driving their unique nuclear, mitochondrial, and biogeographic histories.
鼠耳蝠迅速分化成100多个物种，这是可供研究的最广泛的哺乳动物辐射之一。了解鼠耳蝠内部关系的努力主要利用了线粒体标记，而从核标记推断出的系统发育树缺乏分辨率。因此，我们目前对鼠耳蝠内部关系的理解偏向于一组可能无法反映核基因组历史的系统发育标记。为了解决这个问题，我们对37种有代表性的鼠耳蝠（主要来自新大陆）的完整线粒体基因组进行了测序，并结合对3648个超保守元件（UCE）的靶向测序。我们推断了系统发育，并探讨了串联和总结系统发育方法以及基于信息性或缺失数据水平的标记组合对我们结果的影响。从核UCE数据生成的294个系统发育树中，所有这些树都与使用线粒体基因组推断出的系统发育树有显著差异。即使在核数据中，四重奏频率表明，所有UCE位点中约有一半与估计的物种树冲突。几个因素可能导致这种冲突，包括不完全谱系分选、渐渗杂交，甚至系统发育错误。尽管核UCE位点与线粒体基因组之间以及UCE位点本身之间存在不一致程度，但在所有分析中有四分之一的分析中恢复了最常见的核拓扑结构，并得到了强烈的节点支持。基于这些结果，我们重新审视了鼠耳蝠的进化历史，以更好地理解驱动其独特的核、线粒体和生物地理历史的现象。
