Articles of Significant Interest In This Issue

Abstract

Genomic epidemiology studies identify single nucleotide polymorphisms (SNPs). Methods to infer selection using coding sequences are designed for fixed differences rather than polymorphisms. These methods are therefore not suited to outbreak and chronic infection data, in which SNPs are not necessarily fixed. Using models, Loo et al. (e02002-20) quantified the proportion of nonsynonymous SNPs under selection. They developed an approach for comparing SNP data to the neutral expectation and applied these methods to simulation and empirical data. They found that positive selection is generally difficult to detect, but their analysis suggests different underlying mechanisms of evolution in within-host versus between-host data.