基于转录组挖掘的桫椤目系统发育基因组学研究

This project aims for reconstructing a comprehensive and well-supported phylogeny of Cyatheales based on transcriptome-mining for single-copy nuclear (SCN) markers of representative species of Cyatheales and its related groups. The main research contents are as follows: (1) We will use three strategies—Second Generation Sequencing (SGS), Third Generation Sequencing (TGS), and hybrid sequencing (“Hybrid-Seq”), which integrates the former two technologies—to sequence the transcriptome of representative species of Cyatheales and its related groups. Reads will be assembled, annotated, and used to perform routine transcriptomics analysis. The overall performance of the three strategies will also be compared and evaluated. (2) We will mine the transcriptome data to identify orthologs and in-paralogs (paralogs that arose from a duplication event after the speciation), by clustering which taxon-specific SCN makers can be developed for Cyatheales with the curation of a tree-searching approach. Primers will be designed and tested their relative success in amplifying the desired SCN region from genomic DNA. A test set of species will also be created to characterize their utility in inferring relationships at various phylogenetic depths. (3) Nucleotide alignments, codon alignments, and amino acid alignments will be constructed from the SCN sequences of Cyatheales. They will further be organized into different models by incorporating varied partitioning schemes. Maximum likelihood (ML) analysis will be conducted to infer the ML tree and to detect the effects of model mis-specifications. Bayesian approach will be applied to infer a time-calibrated phylogeny, using a relaxed clock model. And a coalescent-based algorithm will be used to estimate the species tree, to explore the causes for conflicting topologies between gene trees and species trees, and to dissect the effects of incomplete lineage sorting on species tree building. Phylogenomic results of this project may provide valuable evidences for resolving difficult phylogenetic questions of Cyatheales.

本项目拟基于转录组挖掘开展桫椤目的系统发育基因组学研究。内容包括：一、选取桫椤目及其相关类群的代表植物，采用二代、三代及二代和三代技术的联合（Hybrid-Seq）等策略对其转录组进行测序、组装和转录组学分析；并对三种策略进行比较和评估；二、由转录组数据挖掘直系同源物和In-paralog，对其进行聚类和构树搜寻，以在基因组水平开发桫椤目的单拷贝核基因标记；设计引物，检测其扩增目标核基因的有效性；组建测试物种集，验证基因的系统发育可用性；三、对单拷贝核基因序列，从核苷酸、密码子及氨基酸等层次构建数据矩阵和设置分划方案；然后，使用极大似然法重建系统树，研究进化模型的设定对构树产生的效应；借助贝叶斯途径，基于放松分子钟模型推测系统发育，估算支系的分歧时间；利用基于溯祖的算法，研究基因树和物种树之间的冲突，解析谱系分选对物种树构建造成的影响。最后，综合系统发育基因组学的结果，探究疑难系统学问题。

本项目采用高通量测序技术测得了桫椤目及其相关类群的转录组序列数据，然后基于转录组挖掘开展了系统发育基因组学研究。项目解析了影响系统发育树重建的关键因素，并综合系统发育基因组学的结果，对桫椤目的一些疑难系统学问题进行了探究。以下为业已完成的研究内容、获得的重要结果和进展：（1）对桫椤目及其相关类群共计44种代表植物，完成了它们转录组的测序和组装。揭示了RNA 质量对转录组组装产生的效应。（2）将二代、三代以及二代和三代技术的联合（也即Hybrid sequencing, Hybrid-Seq）等三种测序策略分别对金毛狗、白桫椤、大叶黑桫椤、黑桫椤及荫生桫椤的转录组进行了分析，并从多个维度对三种策略进行了综合比较和评估。注意到三代测序技术能更好地揭示选择性剪接和基因/基因家族的复杂性，而Hybrid-Seq 则可有效提高数据产出并能更加准确地估测转录本丰度。（3）利用OrthoMCL-DB，鉴定了桫椤目代表植物的直系同源群，并在此基础上确定出单拷贝核基因。（4）基于挖掘出的单拷贝核基因序列数据，重建了桫椤目的系统发育关系，估算了支系的分歧时间。明确了GC偏斜式样、密码子偏好性、以及SSR（Simple sequence repeat）和转座元件（Transposable element）的类型与分布等分子性状的系统发育价值。根据由这些新型分子性状获得的系统发育基因组学研究结果，对一些疑难系统学问题予以了探究。（5）利用全长转录组数据，研究了基因表达水平和密码子使用偏好对蛋白质编码基因序列的进化式样带来的效应，以及由此对基于序列进化模型重建系统发育树产生的影响。项目的研究结果发表在Journal of Proteomics、BMC Plant Biology、BMC Genomics、Ecology and Evolution及Plant Systematics and Evolution等重要SCI国际学术刊物，被国内外其他研究组多次引用。