soft sweep

Glémin S & Ronfort J 2013 Adaptation and maladaptation in selfing and outcrossing species: new mutations versus standing variation. Evolution 67:225-240.

• simulations
• to compute the probability of fixation from standing genetic variation, simulations are started 6N generations before the environmental change as in (Hermisson and Pennings 2005) to let the population reach mutation-drift or mutation-selection-drift balance
• after 6N generations, new mutational input is stopped

• Orr and Betancourt (2001) showed that P_sv is highest for very recessive alleles, the opposite of Haldane's sieve, because their initial frequency more than compensate the lower selection they experience after the environmental change

• except for specific conditions (very recessive and weakly advantageous mutations), the probability of fixation from standing variation should be less likely in selfers than in outcrossers
• this is due to two effects of selfing:
• (1) lower levels of polymorphism are maintained in selfers because N_e is reduced
• (2) initially deleterious alleles are maintained at lower frequencies because of the purging process

• in domesticated plants, we recently showed that QTL fixed during the domestication process (considered as an adaptation process) tends to be more dominant in outcrossers than in selfers (J. Ronfort and S. Glémin, unpubl. ms.), in agreement with Haldane's sieve occurring on new mutations in outcrossers
• some domestication QTLs are recessive in outcrossers, suggesting that the corresponding alleles likely preexist in wild populations
• in maize (an outcrossing species) for instance, several major genes involved in the domestication syndrome, including Tb1, contribute to standing variation in its wild progenitor, Teosinte

• selfing could have facilitated adaptation during domestication by favoring the fixation of recessive, initially strongly deleterious traits, which could contribute to explaining the bias toward selfing among domesticated species