Kostikova Anna, Silvestro Daniele, Pearman Peter B, Salamin Nicolas
Department of Ecology and Evolution, University of Lausanne, 1015 Lausanne, Switzerland; Swiss Institute of Bioinformatics, Quartier Sorge, 1015 Lausanne, Swizterland;
Department of Ecology and Evolution, University of Lausanne, 1015 Lausanne, Switzerland; Swiss Institute of Bioinformatics, Quartier Sorge, 1015 Lausanne, Swizterland; Department of Biological and Environmental Sciences, University of Gothenburg, Carl Skottsbergsgata 22B, 413 19 Gothenburg, Sweden;
Syst Biol. 2016 May;65(3):417-31. doi: 10.1093/sysbio/syw010. Epub 2016 Feb 23.
The evolution of organisms is crucially dependent on the evolution of intraspecific variation. Its interactions with selective agents in the biotic and abiotic environments underlie many processes, such as intraspecific competition, resource partitioning and, eventually, species formation. Nevertheless, comparative models of trait evolution neither allow explicit testing of hypotheses related to the evolution of intraspecific variation nor do they simultaneously estimate rates of trait evolution by accounting for both trait mean and variance. Here, we present a model of phenotypic trait evolution using a hierarchical Bayesian approach that simultaneously incorporates interspecific and intraspecific variation. We assume that species-specific trait means evolve under a simple Brownian motion process, whereas species-specific trait variances are modeled with Brownian or Ornstein-Uhlenbeck processes. After evaluating the power of the method through simulations, we examine whether life-history traits impact evolution of intraspecific variation in the Eriogonoideae (buckwheat family, Polygonaceae). Our model is readily extendible to more complex scenarios of the evolution of inter- and intraspecific variation and presents a step toward more comprehensive comparative models for macroevolutionary studies.
生物体的进化在很大程度上依赖于种内变异的进化。它与生物和非生物环境中的选择因子之间的相互作用是许多过程的基础，如种内竞争、资源分配以及最终的物种形成。然而，性状进化的比较模型既不允许对与种内变异进化相关的假设进行明确检验，也不能通过同时考虑性状均值和方差来估计性状进化速率。在这里，我们提出了一种使用分层贝叶斯方法的表型性状进化模型，该模型同时纳入了种间和种内变异。我们假设特定物种的性状均值在一个简单的布朗运动过程中进化，而特定物种的性状方差则用布朗或奥恩斯坦 - 乌伦贝克过程建模。在通过模拟评估该方法的效力后，我们研究生活史性状是否会影响蓼科荞麦亚科（蓼科）种内变异的进化。我们的模型很容易扩展到种间和种内变异进化的更复杂场景，并朝着宏观进化研究更全面的比较模型迈出了一步。
