Do macrophylogenies yield stable macroevolutionary inferences? An example from squamate reptiles

Advances in the generation, retrieval, and analysis of phylogenetic data have enabled researchers to create phylogenies that contain many thousands of taxa. These “macrophylogenies”—large trees that typically derive from megaphylogeny, supermatrix, or supertree approaches—provide researchers with an unprecedented ability to conduct evolutionary analyses across broad phylogenetic scales. Many studies have now used these phylogenies to explore the dynamics of speciation, extinction, and phenotypic evolution across large swaths of the tree of life. These trees are characterized by substantial phylogenetic uncertainty on multiple levels, and the stability of macroevolutionary inferences from these data sets has not been rigorously explored. As a case study, we tested whether five recently published phylogenies for squamate reptiles—each consisting of more than 4000 species—yield congruent inferences about the processes that underlie variation in species richness across replicate evolutionary radiations of Australian snakes and lizards. We find discordance across the five focal phylogenies with respect to clade age and several diversification rate metrics, and in the effects of clade age on species richness. We also find that crown clade ages reported in the literature on these Australian groups are in conflict with all of the large phylogenies examined. Macrophylogenies offer an unprecedented opportunity to address evolutionary and ecological questions at broad phylogenetic scales, but accurately representing the uncertainty that is inherent to such analyses remains a critical challenge to our field.

系统发育数据的生成、检索与分析技术的迭代进步，使得研究者得以构建包含数千个分类单元（taxa）的系统发育树（phylogeny）。这类被称为"大型系统发育树（macrophylogeny）"的演化树，通常基于巨型系统发育树（megaphylogeny）、超矩阵（supermatrix）或超树（supertree）方法构建，为研究者提供了前所未有的能力，可在宽泛的系统发育尺度下开展演化分析。目前已有诸多研究借助此类系统发育树，探究生命之树广阔分支中物种形成、灭绝与表型演化的动态过程。

此类系统发育树在多个层面均存在显著的系统发育不确定性，而基于此类数据集得到的宏观演化推断结果的稳定性，尚未得到严谨的探究。作为案例研究，我们针对5篇近期发表的有鳞目爬行动物（squamate reptiles）系统发育树开展检验——每棵树均包含超过4000个物种——旨在验证这些树能否对澳大利亚蛇类与蜥蜴的重复演化辐射事件中，物种丰富度差异背后的驱动过程得出一致的推断结果。我们发现，这5棵目标系统发育树在演化支（clade）年龄、多项分化速率指标以及演化支年龄对物种丰富度的影响方面均存在不一致性。同时我们还发现，现有文献中针对这些澳大利亚类群报道的冠群演化支（crown clade）年龄，与本次检验的所有大型系统发育树均存在冲突。

大型系统发育树为在宽泛系统发育尺度下解答演化与生态学问题提供了前所未有的契机，但精准呈现此类分析固有的不确定性，仍是本领域亟待攻克的关键挑战。