Longo S J, Faircloth B C, Meyer A, Westneat M W, Alfaro M E, Wainwright P C
Department of Evolution and Ecology, University of California, Davis, CA 95616, USA.
Department of Biological Sciences and Museum of Natural Science, Louisiana State University, Baton Rouge, LA 70803, USA.
Mol Phylogenet Evol. 2017 Aug;113:33-48. doi: 10.1016/j.ympev.2017.05.002. Epub 2017 May 6.
Phylogenetics is undergoing a revolution as large-scale molecular datasets reveal unexpected but repeatable rearrangements of clades that were previously thought to be disparate lineages. One of the most unusual clades of fishes that has been found using large-scale molecular datasets is an expanded Syngnathiformes including traditional long-snouted syngnathiform lineages (Aulostomidae, Centriscidae, Fistulariidae, Solenostomidae, Syngnathidae), as well as a diverse set of largely benthic-associated fishes (Callionymoidei, Dactylopteridae, Mullidae, Pegasidae) that were previously dispersed across three orders. The monophyly of this surprising clade of fishes has been upheld by recent studies utilizing both nuclear and mitogenomic data, but the relationships among major lineages within Syngnathiformes remain ambiguous; previous analyses have inconsistent topologies and are plagued by low support at deep divergences between the major lineages. In this study, we use a dataset of ultraconserved elements (UCEs) to conduct the first phylogenomic study of Syngnathiformes. UCEs have been effective markers for resolving deep phylogenetic relationships in fishes and, combined with increased taxon sampling, we expected UCEs to resolve problematic syngnathiform relationships. Overall, UCEs were effective at resolving relationships within Syngnathiformes at a range of evolutionary timescales. We find consistent support for the monophyly of traditional long-snouted syngnathiform lineages (Aulostomidae, Centriscidae, Fistulariidae, Solenostomidae, Syngnathidae), which better agrees with morphological hypotheses than previously published topologies from molecular data. This result was supported by all Bayesian and maximum likelihood analyses, was robust to differences in matrix completeness and potential sources of bias, and was highly supported in coalescent-based analyses in ASTRAL when matrices were filtered to contain the most phylogenetically informative loci. While Bayesian and maximum likelihood analyses found support for a benthic-associated clade (Callionymidae, Dactylopteridae, Mullidae, and Pegasidae) as sister to the long-snouted clade, this result was not replicated in the ASTRAL analyses. The base of our phylogeny is characterized by short internodes separating major syngnathiform lineages and is consistent with the hypothesis of an ancient rapid radiation at the base of Syngnathiformes. Syngnathiformes therefore present an exciting opportunity to study patterns of morphological variation and functional innovation arising from rapid but ancient radiation.
随着大规模分子数据集揭示出以前被认为是不同谱系的分支出乎意料但可重复的重排,系统发育学正在经历一场革命。利用大规模分子数据集发现的最不寻常的鱼类分支之一是一个扩展的海龙目,包括传统的长吻海龙目谱系(管口鱼科、剃刀鱼科、烟管鱼科、裸胸鳝科、海龙科),以及一组多样的主要与底栖生物相关的鱼类(线鳍鱼亚目、飞角鱼科、羊鱼科、海蛾鱼科),这些鱼类以前分散在三个目中。最近利用核数据和线粒体基因组数据的研究支持了这个令人惊讶的鱼类分支的单系性,但海龙目内主要谱系之间的关系仍然不明确;以前的分析拓扑结构不一致,并且在主要谱系之间的深度分歧处支持率较低。在这项研究中,我们使用超保守元件(UCE)数据集对海龙目进行了首次系统发育基因组学研究。UCE是解决鱼类深度系统发育关系的有效标记,结合增加的分类群采样,我们期望UCE能解决有问题的海龙目关系。总体而言,UCE在一系列进化时间尺度上有效地解决了海龙目内的关系。我们发现传统长吻海龙目谱系(管口鱼科、剃刀鱼科、烟管鱼科、裸胸鳝科、海龙科)的单系性得到了一致支持,这比以前发表的分子数据拓扑结构更符合形态学假设。所有贝叶斯分析和最大似然分析都支持这一结果,对矩阵完整性和潜在偏差来源的差异具有鲁棒性,并且在ASTRAL中基于合并的分析中,当矩阵被过滤以包含最具系统发育信息的位点时,得到了高度支持。虽然贝叶斯分析和最大似然分析支持一个与底栖生物相关的分支(线鳍鱼科、飞角鱼科、羊鱼科和海蛾鱼科)作为长吻分支的姐妹分支,但这一结果在ASTRAL分析中并未得到重复。我们系统发育树的基部特征是主要海龙目谱系之间的节间较短,这与海龙目基部古老的快速辐射假说一致。因此,海龙目为研究快速但古老的辐射产生的形态变异模式和功能创新提供了一个令人兴奋的机会。