Department of Biology at Texas A&M University, College Station, TX 77840, United States; Ecology & Evolutionary Biology IDP at Texas A&M University, College Station, TX 77840, United States; Department of Integrative Biology, Michigan State University, 293 Farm Ln., East Lansing, MI 48825, United States.
Department of Biology at Texas A&M University, College Station, TX 77840, United States; Ecology & Evolutionary Biology IDP at Texas A&M University, College Station, TX 77840, United States.
Mol Phylogenet Evol. 2022 Aug;173:107505. doi: 10.1016/j.ympev.2022.107505. Epub 2022 May 14.
The tendency to discretize biology permeates taxonomy and systematics, leading to models that simplify the often continuous nature of populations. Even when the assumption of panmixia is relaxed, most models still assume some degree of discrete structure. The multispecies coalescent has emerged as a powerful model in phylogenetics, but in its common implementation is entirely space-independent - what we call the "missing z-axis". In this article, we review the many lines of evidence for how continuous spatial structure can impact phylogenetic inference. We illustrate and expand on these by using complex continuous-space demographic models that include distinct modes of speciation. We find that the impact of spatial structure permeates all aspects of phylogenetic inference, including gene tree stoichiometry, topological and branch-length variance, network estimation, and species delimitation. We conclude by utilizing our results to suggest how researchers can identify spatial structure in phylogenetic datasets.
生物离散化的趋势渗透到分类学和系统学中，导致模型简化了种群中经常出现的连续性质。即使放宽了泛群集的假设，大多数模型仍然假设存在某种程度的离散结构。多物种合并已成为系统发育学中的强大模型，但在其常见实现中完全是与空间无关的 - 我们称之为“缺失 z 轴”。在本文中，我们回顾了许多有关连续空间结构如何影响系统发育推断的证据。我们通过使用包括不同物种形成模式的复杂连续空间人口模型来说明和扩展这些证据。我们发现，空间结构的影响渗透到系统发育推断的所有方面，包括基因树化学计量、拓扑和分支长度方差、网络估计和物种划分。最后，我们利用我们的结果提出了研究人员如何在系统发育数据集识别空间结构的建议。
