4 - Agrogenom species phylogeny

Different methods were attempted to estimate the species tree:a) concatenating gene alignements and using RAxML (under a GTRCAT model, 50 rate categories) to infer the tree from the single concatenate: 1) of 455 core unicopy gene families present in all 47 genomes; 2) of 49 unicopy ribosomal gene families present in at least 45/47 genomes;b) using multiple concatenate of a random jackknife sample of genes and inferring a tree with PhyML (under a GTR+G8+I model) for each of them, and then generate the MRE consensus and use split frequency from the sample's trees to infer branch support: 3) from 500 samples of 25 gene families out of the 455 unicopy core families.<br><br>The reftree/ folder contains the chosen phylogeny, which is based on strategy (3) and further tree annotations such as internal node names.

本研究采用多种方法估算物种树：
a) 将基因序列比对结果进行拼接，基于GTRCAT模型（设置50个速率类别），通过RAxML从单次拼接数据集推断物种树：1）使用所有47个基因组中均存在的455个核心单拷贝基因家族（core unicopy gene families）；2）使用至少在45/47个基因组中存在的49个单拷贝核糖体基因家族（unicopy ribosomal gene families）；
b) 对基因进行随机折刀抽样（random jackknife sample）并构建多组拼接数据集，针对每组数据集基于GTR+G8+I模型通过PhyML推断物种树，随后生成MRE一致性树（MRE consensus tree），并利用抽样得到的各树的分裂频率（split frequency）计算分支支持度（branch support）：3）从455个核心单拷贝基因家族中抽取25个基因家族，共进行500次抽样。

reftree/ 文件夹中包含基于策略3选定的最终系统发育树，以及该树的附加注释信息（如内部节点名称等）。