Comparative analyses of species delimitation methods with molecular data in snappers (Perciformes: Lutjaninae)

The integration of approaches that allow the incorporation of stochasticity of gene histories with phylogenetic methods resulted in new approaches for the old issue of species delimitation. Nevertheless, coalescent methods seem problematic for taxa with large effective population size and shallow temporal diversification (like marine fishes). Here, we investigate the performance of single-locus (cytochrome oxidase 1, commonly used in DNA barcoding initiatives) methods for molecular species delimitation in snappers of Lutjaninae from the Western Atlantic and Pacific Eastern. Our results show incongruences among methods. ABGD, PTP and mPTP trend towards a lower number of estimated species. Phylogenetic-coalescent methods with single threshold were majority congruent for a same number of lineages. On the other hand, algorithms with multiple thresholds tend to estimate a higher number of potential species. We do not endorse the use of single-locus for species delimitation, but we do reinforce that single-locus data is sufficient to flag many problems.

将整合基因历史随机性的方法与系统发育分析方法相结合，为物种界定这一经典难题带来了全新的研究路径。不过，溯祖（coalescent）方法对于有效种群规模庞大、物种分化时间较浅的类群（如海洋鱼类）而言，似乎存在应用局限。本研究以西大西洋及东太平洋地区笛鲷亚科（Lutjaninae）的笛鲷类群为研究对象，探讨单基因座（细胞色素氧化酶1，cytochrome oxidase 1，广泛应用于DNA条形码（DNA barcoding）项目）分子物种界定方法的应用效果。研究结果显示不同方法间存在估算结果不一致的问题：ABGD、PTP及mPTP方法倾向于估算得到较少的物种数量；采用单一阈值的系统发育-溯祖方法，多数情况下对谱系数量的估算结果保持一致。与之相反，采用多阈值的算法则倾向于估算得到更多的潜在物种。本研究不建议将单基因座方法用于物种界定，但同时强调，单基因座数据足以揭示诸多物种界定相关的问题。