Biologists have long been interested in sexually antagonistic selection, in which the genetic variants that provide an advantage for one sex are disadvantageous for the other. Sexual antagonism is important because it helps maintain genetic variation and represents one of several ways in which populations might remain maladapted with respect to their environments.
In 1984, a theoretical prediction was proposed by William Rice that said the X chromosome should be a ‘hot spot’ for sexually antagonistic genetic polymorphism. His mathematical models indicated that sexually antagonistic alleles were more likely to remain polymorphic when they were linked to the X chromosome than when they were on other types of chromosomes (i.e., autosomes). Rice’s prediction that polymorphism is easier to maintain on the X chromosome critically depends on the dominance relations between sexually antagonistic alleles. Other researchers have shown that the autosomes are more conducive to maintaining genetic variation under conditions that differ from Rice’s assumptions.
There have been numerous empirical studies that have demonstrated apparent support for Rice’s theory. But Filip Ruzicka and Tim Connallon argue that these studies share a common but incorrect assumption: that signals from sexually antagonistic genetic variation are equally detectable whether the variation is on sex chromosomes (ie X-linked) or on autosomes.
Instead, Filip and Tim found that the existing methods for testing this classic theory are all biased towards finding signals of sexually antagonistic variation on the X chromosome. They developed mathematical models to test how much X chromosomes and autosomes contributed to signals of sexual antagonism and found a considerable bias in existing studies.
When they revisited the experimental studies using their models, they found that most of them were actually consistent with scenarios in which the X chromosome is not a hot spot for sexually antagonistic polymorphism.
So how can we be sure whether William Rice’s theory is correct or not? Filip and Tim concede that experimentally testing this classic theory is difficult. They suggest the most feasible approach is to compare fitness components between fathers and sons (who do not inherit their fathers X chromosome) with fitness components between fathers and daughters (who do). Modern genomic approaches that directly estimate the fitness effects of individual genetic variation (genome-wide association studies or “GWAS” of fitness) are also promising avenues for testing the theory.
Filip and Tim hope that their predictions provide better guidelines for future tests. Their models can be used as baseline expectations against which experimental data can be compared.
While they acknowledge that there are limitations in their models, they maintain further attention to this issue will greatly improve our ability to predict the potential contributions of X-linked and autosomal genes to population genetic diversity and species divergence.