Prevalence of positive selection among nearly neutral amino acid replacements in Drosophila

Abstract

We have estimated the selective effects of amino acid replacements in natural populations by comparing levels of polymorphism in 91 genes in African populations of Drosophila melanogaster with their divergence from Drosophila simulans. The genes include about equal numbers whose level of expression in adults is greater in males, greater in females, or approximately equal in the sexes. Markov chain Monte Carlo methods were used to sample key parameters in the stationary distribution of polymorphism and divergence in a model in which the selective effect of each nonsynonymous mutation is regarded as a random sample from some underlying normal distribution whose mean may differ from one gene to the next. Our analysis suggests that approximately 95% of all nonsynonymous mutations that could contribute to polymorphism or divergence are deleterious, and that the average proportion of deleterious amino acid polymorphisms in samples is approximately 70%. On the other hand, approximately 95% of fixed differences between species are positively selected, although the scaled selection coefficient (N(e)s) is very small. We estimate that approximately 46% of amino acid replacements have N(e)s < 2, approximately 84% have N(e)s < 4, and approximately 99% have N(e)s < 7. Although positive selection among amino acid differences between species seems pervasive, most of the selective effects could be regarded as nearly neutral. There are significant differences in selection between sex-biased and unbiased genes, which relate primarily to the mean of the distributions of mutational effects and the fraction of slightly deleterious and weakly beneficial mutations that are fixed.

Fig. 1.

Inferred distribution of scaled selection coefficients N_es among new nonsynonymous mutations that could plausibly become polymorphic or fixed, where N_e is the diploid effective population number and s is the conventional selection coefficient. N_es corresponds to the parameter γ/2 in Eq. 1. The mutational distributions exclude all mutations that are lethal or sterile and those with selective effects that are so deleterious as to preclude their becoming polymorphic. The distributions are based on an analysis of 33 genes whose expression is male-biased (red), 28 genes whose expression is female-biased (green), and 30 genes with approximately equal expression in adults of both sexes (blue).

Fig. 2.

Estimated proportion of positively selected nonsynonymous mutations among new mutations (N), sample polymorphisms (S), and fixed differences (F) between D. melanogaster and D. simulans. New mutations include only those that could plausibly become polymorphic or fixed. The error bars are the 95% credible intervals around the means. Blue bars indicate genes expressed approximately equally in adults of both sexes; green bars indicate genes with female-biased expression; red bars indicate genes with male-biased expession; and gold bars indicate all genes combined.

Fig. 3.

Estimated proportion of fixed amino acid replacements between D. melanogaster and D. simulans whose scaled selection coefficient is less than various specified valus of N_es, ordered by rank among all 91 genes. N_e is the diploid effective population number and s is the conventional selection coefficient. Blue dots indicate genes expressed approximately equally in adults of both sexes; green dots indicate genes with female-biased expression; and red dots indicate genes with male-biased expression.

Fig. 4.

Inferred cumulative density function (CDF) of the scaled selection coefficients among fixed amino acid replacements in 91 genes between D. melanogaster and D. simulans. N_e is the diploid effective population number, and s is the conventional selection coefficient. CDF(N_es) is the average proportion of amino acid differences whose scaled selection coefficient is smaller than N_es, and α(N_es) is the proportion of amino acid differences whose scaled selection coefficient is greater than N_es.