School of Life Sciences, Fudan University, Shanghai 200433, China.
Nature. 2010 Dec 16;468(7326):E9; discussion E10. doi: 10.1038/nature09482.
The question of whether or not all life on Earth shares a single common ancestor has been a central problem of evolutionary biology since Darwin. Although the theory of universal common ancestry (UCA) has gathered a compelling list of circumstantial evidence, as given in ref. 2, there has been no attempt to test statistically the UCA hypothesis among the three domains of life (eubacteria, archaebacteria and eukaryotes) by using molecular sequences. Theobald recently challenged this problem with a formal statistical test, and concluded that the UCA hypothesis holds. Although his attempt is the first step towards establishing the UCA theory with a solid statistical basis, we think that the test of Theobald is not sufficient enough to reject the alternative hypothesis of the separate origins of life, despite the Akaike information criterion (AIC) of model selection giving a clear distinction between the competing hypotheses.
地球上所有生命是否共享一个单一的共同祖先，这个问题自达尔文以来一直是进化生物学的核心问题。尽管普遍共同祖先（UCA）理论已经收集了一系列有说服力的间接证据，如参考文献 2 所示，但在使用分子序列对生命的三个领域（真细菌、古细菌和真核生物）进行 UCA 假说的统计检验方面，尚未进行过尝试。Theobald 最近通过正式的统计检验挑战了这个问题，并得出结论认为 UCA 假说成立。尽管他的尝试是用坚实的统计基础建立 UCA 理论的第一步，但我们认为，尽管模型选择的 Akaike 信息准则（AIC）在竞争假设之间做出了明确区分，但 Theobald 的检验不足以拒绝生命起源独立的替代假设。
