[Paper Review] The limits of selection under plant domestication

Robin G. Allaby, Dorian Q. Fuller+1 more|arXiv (Cornell University)|Mar 5, 2014

Genetic diversity and population structure48 references3 citations

TL;DR

This study uses population genetic simulations to explore the genetic limits of human-driven selection during plant domestication, demonstrating that selection on 50–100 loci represents an optimal balance between selection load and evolutionary capacity. It reveals that co-selection accelerates fixation in outcrossing species but causes interference in selfing systems, and that selective sweeps are costly in terms of adaptive potential, explaining their rarity in genomic data.

ABSTRACT

Plant domestication involved a process of selection through human agency of a series of traits collectively termed the domestication syndrome. Current debate concerns the pace at which domesticated plants emerged from cultivated wild populations and how many genes were involved. Here we present simulations that test how many genes could have been involved by considering the cost of selection. We demonstrate the selection load that can be endured by populations increases with decreasing selection coefficients and greater numbers of loci down to values of about s = 0.005, causing a driving force that increases the number of loci under selection. As the number of loci under selection increases, an effect of co-selection increases resulting in individual unlinked loci being fixed more rapidly in out-crossing populations, representing a second driving force to increase the number of loci under selection. In inbreeding systems co-selection results in interference and reduced rates of fixation but does not reduce the size of the selection load that can be endured. These driving forces result in an optimum pace of genome evolution in which 50-100 loci are the most that could be under selection in a cultivation regime. Furthermore, the simulations do not preclude the existence of selective sweeps but demonstrate that they come at a cost of the selection load that can be endured and consequently a reduction of the capacity of plants to adapt to new environments, which may contribute to the explanation of why selective sweeps have been so rarely detected in genome studies.

Motivation & Objective

• To investigate the upper limit of genetic loci that could be under selection during plant domestication without exceeding tolerable selection load.
• To examine how co-selection and selection coefficient magnitude influence the rate of fixation in outcrossing versus selfing plant systems.
• To assess whether the rarity of detected selective sweeps in plant genomes can be explained by the evolutionary cost of such sweeps.
• To determine the optimal pace of genome evolution under human-directed selection in domestication contexts.

Proposed method

• Simulated population genetic models to assess selection load under varying numbers of loci and selection coefficients (s).
• Modeled both outcrossing and selfing reproductive systems to compare fixation dynamics under co-selection.
• Tracked fixation rates of unlinked loci under different selection regimes to quantify the impact of co-selection.
• Varied selection coefficients down to s = 0.005 to assess the threshold at which selection load becomes sustainable.
• Evaluated the trade-off between rapid fixation and the capacity to adapt to new environments.
• Used computational simulations to test whether selective sweeps could occur without compromising long-term evolutionary flexibility.

Experimental results

Research questions

• RQ1 What is the maximum number of loci that could be under selection during plant domestication without exceeding tolerable selection load?
• RQ2 How does co-selection affect the rate of fixation in outcrossing versus selfing plant populations?
• RQ3 Why are selective sweeps so rarely detected in genomic studies of domesticated plants?
• RQ4 Does the cost of maintaining a high selection load limit the evolutionary capacity of domesticated plants?
• RQ5 What is the optimal number of loci under selection that balances rapid domestication with long-term adaptability?

Key findings

• The maximum number of loci under selection in a cultivation regime is estimated to be between 50 and 100, representing an optimal balance of selection load and evolutionary capacity.
• Selection load decreases with lower selection coefficients (s ≈ 0.005) and increasing numbers of loci, creating a driving force for more loci to be under selection.
• In outcrossing systems, co-selection accelerates fixation of unlinked loci, increasing the efficiency of selection.
• In selfing systems, co-selection leads to interference and reduced fixation rates, but does not reduce the total selection load the population can endure.
• Selective sweeps are evolutionarily costly, reducing the population's capacity to adapt to new environments, which may explain their low detection frequency in genomic data.
• The simulations do not rule out selective sweeps but show they come at the expense of long-term adaptability, suggesting a trade-off between rapid domestication and evolutionary resilience.