Predicting Multivariate Responses of Sexual Dimorphism to Direct and Indirect Selection

Abstract

Sexual dimorphism is often assumed to result from balancing the strength of antagonistic selection in favor of dimorphism against the degree of constraint imposed by the shared genome of the sexes, reflected in the B matrix of genetic intersexual covariances. To investigate the totality of forces shaping dimorphism, we reparameterized the Lande equation to predict changes in trait averages and trait differences between the sexes. As genetic constraints on the evolution of dimorphism in response to antagonistic selection become larger, dimorphism will tend to respond more rapidly to concordant selection (which favors the same direction of change in male and female traits) than to antagonistic selection. When we apply this theory to four empirical estimates of B in Drosophila melanogaster, the indirect responses of dimorphism to concordant selection are of comparable or larger magnitude than the direct responses of dimorphism to antagonistic selection in two suites of traits with typical levels of intersex correlation. Antagonistic selection is more important in two suites of traits where the intersex correlations are unusually low. This suggests that the evolution of sexual dimorphism may sometimes be dominated by concordant selection rather than antagonistic selection.

The genetic architecture of sexual dimorphism in the mossCeratodon purpureus

“…Intuitively, it would make sense that the similarity between G f and G m would impose a genetic constraint. However, Cheng & Houle [20] demonstrated that similarity in male and female covariance matrices coupled with some degree of B matrix asymmetry suggests a greater opportunity for sexual dimorphism in response to sexually concordant selection than to sexually antagonistic selection. Thus, our estimates of the proportion of standing genetic variation that could respond to sexually antagonistic selection represent lower bounds for the potential sexually dimorphic response, as further sexual dimorphism could evolve in response to sexually concordant selection.…”

“…An analysis of the degree of multivariate sexually antagonistic genetic variation in G mf provides insight to the capacity for further response to sexually antagonistic selection [20,52]. The overall percentages of sexually antagonistic genetic variance were estimated with wide, highly overlapping CIs between our two trait categories (table 2).…”

“…Theory and empirical evidence show that opposing selection in males and females can maintain genetic variation for fitness [7][8][9][10][11][12][13][14][15]. However, whether the sexual conflict in a population is evolutionarily transient or persistent will depend on both the nature of sex-specific selection and the nature of sex-specific genetic architecture for traits [16][17][18][19][20], the latter of which remains poorly understood, especially in non-model organisms.…”

“…B need not equal B T ). Asymmetries in B may play an important role in the evolution of sexual dimorphism, although the prevalence of such asymmetry is unknown outside of a few model systems [20,29].…”

Taking quantitative genomics into the wild

“…One phenomenon that can maintain genetic variation for fitness-related traits in natural populations is sexual antagonism, where selection favours different alleles within males and females [63]. Yet, the amount and prevalence of sexually antagonistic variation depends on the genetic architecture of dimorphism, which is not well understood [58,64]. Kollar et al [43] used a multivariate approach to examine the genetic basis of sexual dimorphism in a scent-based fertilization syndrome in the moss Ceratodon purpureus.…”