Pupko Tal, Galtier Nicolas
The Institute of Statistical Mathematics, 4-6-7 Minami-Azabu Minato-ku, Tokyo 106-8569, Japan.
Proc Biol Sci. 2002 Jul 7;269(1498):1313-6. doi: 10.1098/rspb.2002.2025.
A new method for detecting site-specific variation of evolutionary rate (the so-called covarion process) from protein sequence data is proposed. It involves comparing the maximum-likelihood estimates of the replacement rate of an amino acid site in distinct subtrees of a large tree. This approach allows detection of covarion at the gene or the amino acid levels. The method is applied to mammalian-mitochondrial-protein sequences. Significant covarion-like evolution is found in the (simian) primate lineage: some amino acid positions are fast-evolving (i.e. unconstrained) in non-primate mammals but slow-evolving (i.e. highly constrained) in primates, and some show the opposite pattern. Our results indicate that the mitochondrial genome of primates reached a new peak of the adaptive landscape through positive selection.
提出了一种从蛋白质序列数据中检测进化速率位点特异性变异(即所谓的共变过程)的新方法。该方法涉及比较大树不同子树中氨基酸位点替换率的最大似然估计值。这种方法能够在基因或氨基酸水平上检测共变。该方法应用于哺乳动物线粒体蛋白质序列。在(猿猴)灵长类谱系中发现了显著的类似共变的进化现象:一些氨基酸位置在非灵长类哺乳动物中进化迅速(即不受约束),而在灵长类动物中进化缓慢(即高度受限),还有一些则呈现相反的模式。我们的结果表明,灵长类动物的线粒体基因组通过正选择达到了适应景观的一个新高峰。