Fig. 2

Potential genotype-phenotype links for a hypothetical target gene. The relationship between strength of molecular adaptation (relative rate of protein-coding changes, dN given dS) are shown with magnitude of phenotype in pink and with rate of phenotypic evolution in blue. Crossed lines indicate an identical interpretation regardless of the direction. Where no association is found for the phenotype but is positive for rate (a), genetic selection has driven rapid (i.e. adaptive) phenotypic change in both directions during evolution (e.g. ASPM and primate brain size). If there is also an association with phenotype (b), interpretation is identical to (a), except the adaptation has been consistently directional. Where an association is not found for the phenotype but is negative for rate (c), molecular adaptation is explicitly associated with slower rates of phenotypic change—such genes can be thought of as phenotype moderators acting to minimise change. If there is an association also found for the phenotype (d), it means that although the gene is acting as a moderator, where molecular adaptation does occur it tends to have been in a consistent direction. Finally, where an association is found only for the phenotype and not the rate (e), it implies that this gene has driven directional phenotypic evolution, but phenotypic change has not occurred rapidly and thus may be associated with selection on another, associated phenotype