Plastid phylogenomics of water lilies, with a focus on partitioning strategies and the development of bioinformatic tools

睡莲和近缘植物形成了一个水生的根茎草本植物分支，代表了开花植物中最早的分化谱系之一。最近，首次对该分支进行了叶绿体系统学研究，表明睡莲在目前的范围内是异源的。此外，它还表明，推断的关系取决于所采用的特定数据分区方案。在系统发育学分析中，最优数据分区的应用以及这些分区的最佳核苷酸替换模型的应用是一个关键方面。目前用于确定最佳分配方案和最佳匹配核苷酸替代模型的软件工具在数据格式和分析算法方面没有完全标准化。因此，不同的软件工具确定不同的分区集和替换模型对于当前睡莲的叶绿体基因组数据集是最优的。目前尚不清楚这些差异是通过系统发育推断导致显著不同的树木拓扑结构，还是有限的分类单元采样的结果。拟议的研究项目旨在生成一个包括一半以上所描述的睡莲及其近缘物种的叶绿体基因组数据集。基于如此广泛的分类单元样本，该项目旨在评估睡莲的进化史以及不同数据划分方案对其系统发育推断的影响。此外，拟议的调查旨在评估软件选择对最优数据划分方案和核苷酸替换模型推断的影响。为促进这项评估，在编写本提案时开发了一个新的软件脚本，使分区查找和模型测试的不同软件工具的应用自动化和标准化。基于这个软件脚本的应用，在不同的数据分区方案的背景下，重新评估了睡莲的单一性。该项目完成后，科学界将更好地了解(A)睡莲在其当前范围内的单系性，(B)系统发育树推断对完整的叶绿体基因组序列数据分区的敏感性，以及(C)在推断最佳数据分区和最佳匹配核苷酸替代模型期间软件选择的影响。拟议项目的结果是深远的，可能会影响我们对被子植物早期进化的整体解释。

The water lilies and relatives form a clade of aquatic, rhizomatous herbs that represents one of the earliest diverging lineages of flowering plants. Recently, the first plastid phylogenomic investigation of this clade indicated that the water lilies were paraphyletic in their current circumscription. Moreover, it indicated that the inferred relationships were dependent on the specific data partitioning scheme employed. In phylogenomic analysis, the application of optimal data partitions as well as best-fitting nucleotide substitution models for these partitions is a critical aspect. Current software tools for the identification of optimal partitioning schemes and best-fitting nucleotide substitution models are not fully standardized regarding data format and analysis algorithm. As a consequence, different software tools identify different partition sets and substitution models as optimal for the current plastid genome dataset of the water lilies. It is currently unknown if these differences lead to significantly different tree topologies upon phylogeny inference or if they are the consequence of a limited taxon sampling. The proposed research project aims to generate a plastid genome dataset that includes more than half of all described species of water lilies and relatives. Based on such a broad taxon sampling, the project aims to evaluate the evolutionary history of the water lilies and the impact of different data partition schemes on their phylogeny inference. Moreover, the proposed investigation aims to assess the impact of software choice on the inference of optimal data partition schemes and nucleotide substitution models. To facilitate this assessment, a novel software script was developed in preparation of this proposal which automates and standardizes the application of different software tools for partition-finding and model-testing. Based on the application of this software script, the monophyly of the water lilies is re-evaluated in the context of different data partition schemes. Upon completion of this project, the scientific community will have a better understanding of (a) the monophyly of the water lilies in their current circumscription, (b) the sensitivity of phylogenetic tree inference to data partitioning of complete plastid genome sequences, and (c) the effect of software choice during inferences of optimal data partitions and best-fitting nucleotide substitution models. The results of the proposed project are far-reaching and may affect our interpretation of early angiosperm evolution as a whole.