Katsura Yukako, Stanley Craig E, Kumar Sudhir, Nei Masatoshi
Department of Biology and Institute for Genomics and Evolutionary Medicine, Temple University, Philadelphia, PA.
Department of Biology and Institute of Molecular Evolutionary Genetics, Pennsylvania State University, State College, PA.
Mol Biol Evol. 2017 Mar 1;34(3):718-723. doi: 10.1093/molbev/msw272.
The reliability of a phylogenetic tree obtained from empirical data is usually measured by the bootstrap probability (Pb) of interior branches of the tree. If the bootstrap probability is high for most branches, the tree is considered to be reliable. If some interior branches show relatively low bootstrap probabilities, we are not sure that the inferred tree is really reliable. Here, we propose another quantity measuring the reliability of the tree called the stability of a subtree. This quantity refers to the probability of obtaining a subtree (Ps) of an inferred tree obtained. We then show that if the tree is to be reliable, both Pb and Ps must be high. We also show that Ps is given by a bootstrap probability of the subtree with the closest outgroup sequence, and computer program RESTA for computing the Pb and Ps values will be presented.
从经验数据获得的系统发育树的可靠性通常通过树内部分支的自展概率（Pb）来衡量。如果大多数分支的自展概率较高，则该树被认为是可靠的。如果一些内部分支显示出相对较低的自展概率，我们就不能确定推断出的树是否真的可靠。在此，我们提出另一个衡量树可靠性的量，称为子树稳定性。这个量指的是获得推断树的子树（Ps）的概率。然后我们表明，如果树要可靠，Pb和Ps都必须很高。我们还表明，Ps由具有最接近外群序列的子树的自展概率给出，并且将展示用于计算Pb和Ps值的计算机程序RESTA。
