Soft sweeps and beyond: understanding the patterns and probabilities of selection footprints under rapid adaptation

Abstract

Summary1. The tempo and mode of adaptive evolution determine how natural selection shapes patterns of genetic diversity in DNA polymorphism data. While slow mutation-limited adaptation leads to classical footprints of 'hard' selective sweeps, these patterns are different when adaptation responds quickly to a novel selection pressure, acting either on standing genetic variation or on recurrent new mutation. In the past decade, corresponding footprints of 'soft' selective sweeps have been described both in theoretical models and in empirical data. 2. Here, we summarize the key theoretical concepts and contrast model predictions with observed patterns in Drosophila, humans, and microbes. 3. Evidence in all cases shows that 'soft' patterns of rapid adaptation are frequent. However, theory and data also point to a role of complex adaptive histories in rapid evolution. 4. While existing theory allows for important implications on the tempo and mode of the adaptive process, complex footprints observed in data are, as yet, insufficiently covered by models. They call for in-depth empirical study and further model development.Key-words: adaptation, Drosophila, genetic hitch-hiking, HIV, lactase, malaria, selective sweep Hard and soft selective sweeps

Genomic processes underlying rapid adaptation of a natural Chironomus riparius population to unintendedly applied experimental selection pressures

“…While the evidence for the pervasiveness of rapid phenotypic adaptation is piling up (Padfield, Yvon‐Durocher, Buckling, Jennings, & Yvon‐Durocher, ; Colautti & Lau, ; Hairston, Ellner, Geber, Yoshida, & Fox, ), the understanding of the underlying molecular dynamics and genomic architectures is still rather rudimentary (Hermisson & Pennings, ; Jain & Stephan, ; Messer, Ellner, & Hairston, ). Evolve and Resequence (E&R) studies have emerged as useful tool to infer genomic patterns and processes during rapid phenotypic adaptation (Long, Liti, Luptak, & Tenaillon, ; Schlötterer, Kofler, Versace, Tobler, & Franssen, ).…”

Largely unlinked gene sets targeted by selection for domestication syndrome phenotypes in maize and sorghum

“…An additional potential confounding variable is that the mutational target space – defined as the number of potential mutations at a given locus which produce alleles with the same phenotypic outcome – may well vary between genes with larger and smaller phenotypic effects. A larger mutational target space at a given locus increases the probability that the response to artificial selection during domestication will result from a multiple‐origin soft sweep and reduces the potential to identify the locus as a target of selection using bottom‐up population genetic approaches (Hermisson and Pennings, ). The decreasing cost of whole‐genome sequencing and resequencing, and the large number of different grain crops that have experienced parallel selection for domestication syndrome phenotypes, should enable more rigorous tests of this model in the near future incorporating data from syntenic orthologous genes across many different species.…”

The active spread of adaptive variation for reef resilience

“…Translocation models that simulate genetic rescue demonstrate large improvements in survival probabilities and genetic diversity within 40-50 years of the intervention (Pavlova et al, 2017). If fitness benefits are high (i.e., strong selection coefficients exist), then sweeps can occur in as few as 10-100 generations (Coulson et al, 2017;Ferenci, 2008;Hermisson & Pennings, 2017). If fitness benefits are high (i.e., strong selection coefficients exist), then sweeps can occur in as few as 10-100 generations (Coulson et al, 2017;Ferenci, 2008;Hermisson & Pennings, 2017).…”

“…Corals also have overlapping generations, which may initially decrease spread rates through the dilution of the gene pool with non-PAL carrying individuals, but rates may subsequently increase once multiple generations of PAL containing individuals simultaneously reproduce. Encouragingly, beneficial variants need not reach complete fixation to result in an overall positive impact on the receiving populations(Hermisson & Pennings, 2017). Encouragingly, beneficial variants need not reach complete fixation to result in an overall positive impact on the receiving populations(Hermisson & Pennings, 2017).…”