Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor MI 48109-1079
Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor MI 48109-1079.
Proc Natl Acad Sci U S A. 2017 Feb 14;114(7):1607-1612. doi: 10.1073/pnas.1607921114. Epub 2017 Jan 30.
The multispecies coalescent model underlies many approaches used for species delimitation. In previous work assessing the performance of species delimitation under this model, speciation was treated as an instantaneous event rather than as an extended process involving distinct phases of speciation initiation (structuring) and completion. Here, we use data under simulations that explicitly model speciation as an extended process rather than an instantaneous event and carry out species delimitation inference on these data under the multispecies coalescent. We show that the multispecies coalescent diagnoses genetic structure, not species, and that it does not statistically distinguish structure associated with population isolation vs. species boundaries. Because of the misidentification of population structure as putative species, our work raises questions about the practice of genome-based species discovery, with cascading consequences in other fields. Specifically, all fields that rely on species as units of analysis, from conservation biology to studies of macroevolutionary dynamics, will be impacted by inflated estimates of the number of species, especially as genomic resources provide unprecedented power for detecting increasingly finer-scaled genetic structure under the multispecies coalescent. As such, our work also represents a general call for systematic study to reconsider a reliance on genomic data alone. Until new methods are developed that can discriminate between structure due to population-level processes and that due to species boundaries, genomic-based results should only be considered a hypothesis that requires validation of delimited species with multiple data types, such as phenotypic and ecological information.
多物种合并模型是许多物种划分方法的基础。在之前评估该模型下物种划分性能的工作中，物种形成被视为一个瞬间事件，而不是一个涉及物种形成启动（结构）和完成的不同阶段的扩展过程。在这里，我们使用明确将物种形成建模为一个扩展过程而不是瞬间事件的数据，并在多物种合并模型下对这些数据进行物种划分推断。我们表明，多物种合并模型诊断的是遗传结构，而不是物种，并且它不能从统计学上区分与种群隔离相关的结构与物种边界。由于将种群结构错误地识别为假定的物种，我们的工作引发了关于基于基因组的物种发现实践的问题，这在其他领域产生了级联效应。具体来说，从保护生物学到宏观进化动态研究等所有依赖物种作为分析单位的领域，都会受到物种数量膨胀估计的影响，尤其是随着基因组资源为在多物种合并模型下检测越来越细粒度的遗传结构提供了前所未有的能力。因此，我们的工作也代表了对系统研究的普遍呼吁，需要重新考虑仅依赖基因组数据的情况。在开发出能够区分因种群水平过程引起的结构和因物种边界引起的结构的新方法之前，基于基因组的结果只能被视为一个假设，需要使用多种数据类型（如表型和生态信息）来验证划定的物种。
