Data from: Modern spandrels: the roles of genetic drift, gene flow and natural selection in the formation of parallel clines

Urban environments offer the opportunity to study the role of adaptive and
non-adaptive evolutionary processes on an unprecedented scale. While the
presence of parallel clines in heritable phenotypic traits is often
considered strong evidence for the role of natural selection, non-adaptive
evolutionary processes can also generate clines, and this may be more
likely when traits have a non-additive genetic basis due to epistasis. In
this paper, we use spatially-explicit simulations modelled according to
the cyanogenesis (HCN) polymorphism in white clover (Trifolium repens) to
examine the formation of phenotypic clines along urbanization gradients
under varying levels of drift, gene flow and selection. HCN results from
an epistatic interaction between two Mendelian-inherited loci. Our results
demonstrate that the genetic architecture of this trait makes natural
populations susceptible to decreases in HCN frequencies via drift.
Gradients in the strength of drift across a landscape resulted in
phenotypic clines with lower frequencies of HCN in strongly drifting
populations, giving the misleading appearance of deterministic adaptive
changes in the phenotype. Studies of heritable phenotypic change in urban
populations should generate null models of phenotypic evolution based on
the genetic architecture underlying focal traits prior to invoking
selection’s role in generating adaptive differentiation.

城市环境为我们以前所未有的规模研究适应性与非适应性进化过程的作用提供了绝佳契机。尽管可遗传表型性状中出现平行渐变群（parallel clines）常被视为自然选择发挥作用的有力证据，但非适应性进化过程同样可催生渐变群；当性状因上位性（epistasis）具备非加性遗传基础时，这种情况更易发生。本研究基于白三叶（Trifolium repens）的氰合作用（cyanogenesis, HCN）多态性构建空间显式模拟（spatially-explicit simulations）模型，以此探究不同遗传漂变（drift）、基因流（gene flow）与选择水平下，沿城市化梯度形成的表型渐变群。HCN的产生源于两个孟德尔遗传位点（Mendelian-inherited loci）间的上位性互作。我们的研究结果表明，该性状的遗传架构使自然种群易通过遗传漂变导致HCN频率下降。种群空间分布区的漂变强度梯度会催生表型渐变群：强漂变种群中的HCN频率更低，这会误导性地呈现出表型出现确定性适应性变化的假象。针对城市种群可遗传表型变化的研究，在援引选择在催生适应性分化（adaptive differentiation）中的作用之前，应先基于目标性状的遗传架构构建表型进化零模型（null models）。