melanogaster (22, 23), we found that the effects of BGS can account for only part of the observed relation between ?_S and K _{We projected the common energy off choice to the certainly chosen mutations, and also the ratio of new mutations that are beneficial, for both NS and UTR web sites. Another part of our very own method is that it gives gene sales on SSW additionally the BGS patterns, that has a major affect the fresh new factor quotes. As the we unearthed that the outcome with the D. melanogaster–D. yakuba review had greatest analytical characteristics than those into the D. melanogaster ancestry, most likely because they considering so much more real rates of one’s costs out of adaptive development, we focus attract into previous. From now on, we’re going to consider the two datasets just like the mel-yak and you can mel, respectively.}

_{Empirical Efficiency}

_{We first asked whether there was a relation between ?S and KA for autosomal genes located in regions with normal rates of crossing over, using data on all escort girl Rochester available genes (Materials and Methods, Primary Data Analyses); this is displayed in Fig. 1 and SI Appendix, Fig. S1, for mel-yak and mel, respectively. The Spearman rank correlations for the two datasets were small but significantly negative: mel-yak ? = ?0.082, P < 0.001; mel ? = ?0.077, P < 0.001). After correcting for the covariates described in Materials and Methods (SI Appendix, Table S1), the relations were still significant and negative (? = ?0.129, P < 10 ?16 for mel-yak; ? = –0.130, P < 10 ?16 for mel), with a multiple regression coefficient of –0.052 for mel-yak, and ?0.195 for mel. In contrast, the rank partial correlations between ?A and KA were positive, with ? = 0.412, P < 10 ?16 in mel-yak and ? = 0.428 in mel, as has previously been found (e.g., ref. 16). This result implies that the level of selective constraint on a coding sequence is negatively correlated with its KA value, consistent with the result from the DFE-? analyses described next.}

A beneficial, so SSW consequences need to be invoked

The plot of synonymous diversity (?_S) for genes in a Rwandan population of D. melanogaster against their nonsynonymous divergence from D. yakuba (K_A); ? is the Spearman rank correlation coefficient. The green line is the least-squares linear regression (the dashed lines are its 95% CIs).

To examine the potential contributions of BGS and SSWs to the pattern for ?_S, we applied DFE-? (24) to each of 50 bins of K_A values, assuming gamma distributions of the selection coefficients for deleterious mutations within bins, as described in Materials and Methods, Primary Data Analyses (Table 1 and SI Appendix, Tables S2 and S3). The Spearman rank correlations for the mel-yak comparison across bins for ?_S versus ?_a and ?_S versus ?_na were –0.681 (P = 5 ? 10 ?8 ) and –0.727 (P = 2.26 ? 10 ?9 ). Here, ?_a is the ratio of the rate of substitution of positively selected NS mutations to the rate of substitution of synonymous mutations as measured by the synonymous site divergence K_S (25); ?_na is the corresponding ratio for substitutions of neutral or slightly deleterious NS mutations (26) and is an inverse measure of the level of selective constraint on the protein sequence. For mel, the rank correlations were ?0.829 (P < 10 ?13 ) and ?0.845 (P < 10 ?13 ), respectively. However, because both ?_a and ?_na for NS sites increase across bins of K_A (SI Appendix, Table S2), these results do not distinguish between their respective contributions to the patterns for ?_S. To pursue this question, it is necessary to generate predictions for both the BGS and SSW models. The relevant theory is described in Materials and Methods and SI Appendix, sections 1–6. In the next section, we investigate the effects of BGS alone on the relation between ?_S and ?_na, to examine the extent to which BGS could explain the observed relation between ?_S and K_A.