Snir Sagi, Wolf Yuri I, Koonin Eugene V
Department of Evolutionary and Environmental Biology and The Institute of Evolution, University of Haifa, Israel.
National Center for Biotechnology Information, NLM, National Institutes of Health, Bethesda, MD.
Genome Biol Evol. 2014 May 7;6(6):1268-78. doi: 10.1093/gbe/evu091.
Gene evolution is traditionally considered within the framework of the molecular clock (MC) model whereby each gene is characterized by an approximately constant rate of evolution. Recent comparative analysis of numerous phylogenies of prokaryotic genes has shown that a different model of evolution, denoted the Universal PaceMaker (UPM), which postulates conservation of relative, rather than absolute evolutionary rates, yields a better fit to the phylogenetic data. Here, we show that the UPM model is a better fit than the MC for genome wide sets of phylogenetic trees from six species of Drosophila and nine species of yeast, with extremely high statistical significance. Unlike the prokaryotic phylogenies that include distant organisms and multiple horizontal gene transfers, these are simple data sets that cover groups of closely related organisms and consist of gene trees with the same topology as the species tree. The results indicate that both lineage-specific and gene-specific rates are important in genome evolution but the lineage-specific contribution is greater. Similar to the MC, the gene evolution rates under the UPM are strongly overdispersed, approximately 2-fold compared with the expectation from sampling error alone. However, we show that neither Drosophila nor yeast genes form distinct clusters in the tree space. Thus, the gene-specific deviations from the UPM, although substantial, are uncorrelated and most likely depend on selective factors that are largely unique to individual genes. Thus, the UPM appears to be a key feature of genome evolution across the history of cellular life.
传统上,基因进化是在分子钟(MC)模型的框架内进行考量的,即每个基因都具有大致恒定的进化速率。最近对众多原核基因系统发育树的比较分析表明,一种不同的进化模型,即通用起搏器(UPM)模型,假定相对进化速率而非绝对进化速率的保守性,能更好地拟合系统发育数据。在此,我们表明,对于来自六种果蝇和九种酵母的全基因组系统发育树集合,UPM模型比MC模型拟合得更好,具有极高的统计学显著性。与包含远缘生物和多次水平基因转移的原核生物系统发育树不同,这些是简单的数据集,涵盖密切相关的生物群体,并且由与物种树具有相同拓扑结构的基因树组成。结果表明,谱系特异性和基因特异性速率在基因组进化中都很重要,但谱系特异性贡献更大。与MC模型类似,UPM模型下的基因进化速率高度离散,与仅由抽样误差预期的情况相比,大约高出2倍。然而,我们表明果蝇和酵母基因在树空间中都没有形成明显的聚类。因此,基因特异性与UPM的偏差虽然很大,但不相关,很可能取决于主要是各个基因特有的选择因素。因此,UPM似乎是整个细胞生命历史中基因组进化的一个关键特征。