From microevolutionary processes to macroevolutionary patterns: modelling long-term evolution driven by ecological adaptation

几个世纪以来，长期的进化模式吸引了生物学家的想象力，而理解“生命之树”是科学中的一个基本问题。宏观进化模式，如停滞和适应性辐射已被广泛分析，它通常被认为是生态适应发挥了重要作用的长期进化。然而，宏观进化模式的生态学解释仍然主要基于语言理论。宏观进化模式最终是微观进化过程的结果。通过本文的研究，我打算发展一个宏观进化动力学和系统发育分析的数学理论，这是来自生态的出生-死亡过程。我建议使用这个理论作为现有的比较系统发育方法的试验场。最后，我建议应用新的计算工具来研究大型系统发育树上生态重要性状的宏观进化。具体而言：1）我建议使用适应性动力学模型，包括多样化和灭绝的分类显着模式的宏观进化动力学产生的不同类型的生态相互作用，为不同维度的表型空间，并为不同的假设进化社区的地理。这将产生一个一般分类的类型的宏观进化模式所产生的长期进化的机械模型。特别是，它将使显着的宏观进化言语理论的批判性评价。2)自适应动力学模型产生人工系统发育树，我建议使用各种比较系统发育的方法进行分析，以测试这些可以最好地解释的遗传。这将允许从比较方法中得出的推论与产生生态系统的实际生态过程进行比较。这项工作也将提供重要的见解宏观进化的机械模型的统计特性。3)微生物塑造了地球生态系统的化学，确定原核生物生命树中代谢功能的宏观进化模式是一个重要问题。原始遗传可能非常大，它们的分析需要新的计算工具。我建议将这些工具应用于蛋白质超家族沿着微生物生命之树的进化问题。

Long-term patterns of evolution have captured the imagination of biologists for centuries, and understanding the “tree of life” is a fundamental problem in science. Macroevolutionary patterns such as stasis and adaptive radiations have been extensively analyzed, and it is generally thought that ecological adaptations play an important role for long-term evolution. However, ecological explanations for macroevolutionary patters are still mostly based on verbal theories. Macroevolutionary patterns ultimately result from microevolutionary processes. With the research proposed here, I intend to develop a mathematical theory for macroevolutionary dynamics and phylogenetic analysis that is derived from ecological birth-death processes. I propose to use this theory as a testing ground for existing comparative phylogenetic methods. Finally, I propose to apply novel computational tools to study macroevolution of ecologically important traits on large phylogenetic trees. Specifically: 1) I propose to use adaptive dynamics models incorporating diversification and extinction to classify salient patterns of macroevolutionary dynamics generated by different types of ecological interactions, for varying dimensions of phenotype space, and for different assumptions about the biogeography of the evolving communities. This will yield a general classification of the types of macroevolutionary patterns generated by mechanistic models for long-term evolution. In particular, it will enable a critical evaluation of salient verbal theories for macroevolution. 2) The adaptive dynamics models yield artificial phylogenetic trees that I propose to analyze using a variety of comparative phylogenetic methods in order to test which of these can best explain the phylogenies. This will allow for a comparison of inferences made from comparative methods with the actual ecological processes that generated the phylogenies. This work will also give important insights into the statistical properties of mechanistic models for macroevolution. 3) Microorganisms have shaped the chemistry of Earth's ecosystems, and determining the macroevolutionary patterns of metabolic functions in the prokaryotic tree of life is an important problem. Prokaryotic phylogenies can be extremely large, and their analysis requires novel computational tools. I propose to apply such tools to the problem of the evolution of protein superfamilies along the microbial tree of life.