The Effect of Alignment Uncertainty, Substitution Models and Priors in Building and Dating the Mammal Tree of Life

The CDS sequences derived from Liu et al. (2017) were processed to remove potential low-quality regions and indels. The aligned DNA sequences were trimmed and the sequences corresponding to any unusually long terminal branches were removed. Finally, the first and second codon positions (C12) were extracted from 4 sets of alignments - before trimming (BT), after trimming (AT), filtered by the protocol of Irisarri et al. (FP), and further filtered by comparing ML trees with the reference tree (FM). The extracted C12 positions of the BT, AT, FP, and FM alignment were used here to assess uncertainty of alignments and its effect on gene trees, species trees, and divergence time estimation.

源自Liu等人（2017）研究的编码序列（Coding Sequence, CDS）经预处理以去除潜在的低质量区域及插入缺失（indels）。随后对已完成序列比对的DNA序列进行修剪，并移除对应异常长末端分支的序列。最终，从四组比对序列集中提取第一与第二密码子位点（C12），分别为：修剪前比对集（BT）、修剪后比对集（AT）、按照Irisarri等人的实验流程过滤得到的比对集（FP），以及通过最大似然树（Maximum Likelihood Tree, ML Tree）与参考树比对完成进一步过滤得到的比对集（FM）。本研究使用提取自BT、AT、FP及FM比对集的C12位点，评估序列比对的不确定性，及其对基因树、物种树与分化时间估算的影响。