Codon bias and selection on single genomes

Abstract

Arising from: J. B. Plotkin, J. Dushoff & H. B. Fraser Nature 428, 942–945 (2004); see also communication from Nielsen et al.; Chen et al.; Plotkin et al. reply The idea that natural selection on genes might be detected using only a single genome has been put forward by Plotkin and colleagues¹, who present a method that they claim can detect selection without the need for comparative data and which, if correct, would confer greater power of analysis with less information. Here we argue that their method depends on assumptions that confound their conclusions and that, even if these assumptions were valid, the authors' inferences about adaptive natural selection are unjustified.