LIKELIHOOD RATIO TESTS FOR DETECTING POSITIVE SELECTION AND APPLICATION TO PRIMATE LYSOZYME EVOLUTION

An excess of nonsynonymous substitutions over synonymous ones is an im portant indicator of positive selection at the molecular level. A line age that underwent Darwinian selection may have a nonsynonymous/synony mous rate ratio (d(N)/d(S)) that is different from those of other line ages or greater than one. in this paper, several codon-based likelihoo d models that allow for variable d(N)/d(S) ratios among lineages were developed. They were then used to construct likelihood ratio tests to examine whether the d(N)/d(S) ratio is variable among evolutionary lin eages, whether the ratio for a few lineages of interest is different f rom the background ratio for other lineages in the phylogeny, and whet her the d(N)/d(S) ratio for the lineages of interest is greater than o ne. The tests were applied to the lysozyme genes of 24 primate species . The d(N)/d(S) ratios were found to differ significantly among lineag es, indicating that the evolution of primate lysozymes is episodic, wh ich is incompatible with the neutral theory. Maximum-likelihood estima tes of parameters suggested that about nine nonsynonymous and zero syn onymous nucleotide substitutions occurred in the lineage leading to ho minoids, and the d(N)/d(S) ratio for that lineage is significantly gre ater than one. The corresponding estimates for the lineage ancestral t o colobine monkeys were nine and one, and the d(N)/d(S) ratio for the lineage is not significantly greater than one, although it is signific antly higher than the background ratio. The likelihood analysis thus c onfirmed most, but not all, conclusions Messier and Stewart reached us ing reconstructed ancestral sequences to estimate synonymous and nonsy nonymous rates for different lineages.