deleterious mutation - senecio7の日記

Caballero A, Cusi E, García C & García-Dorado A 2002 Accumulation of deleterious mutations: additional Drosophila melanogaster estimates and a simulation of the effects of selection. Evolution 56:1150-1159.

the simulation results indicated that a model of many mutations of small effect is incompatible with the evolution of the mean viability of the control and MA lines over generations, the distribution of line means after 210 generations of mutation accumulation, and the pattern of line extinction over generations

simulations were run to evaluate the expected behavior of the large control population and the MA lines over the experiment under each assumed model of mutations
a multilocus model with multiplicative fitness action among loci was assumed
every generation, an average number (Poisson distributed) of λ new deleterious mutations per haploid genome arose and were randomly assigned to nonsegregating positions of the genome
the first model was based on the original BM estimates from the quasi-normal chromosomes given by Mukai et al. (1972, p. 350), that is, λ = 0.43 per haploid genome and generation, E(s) = 0.026
the shape parameter for the gamma distribution of homozygous effects was set to 0.51

the second model, which we will refer to as the MD-Mukai model, was based on analysis performed on the same data (García-Dorado et al. 1998) and implies that a part of the fitness reduction observed by Mukai et al. (1972) was not mutational
for this model, λ = 0.011, E(s) = 0.191, and the gamma shape parameter is 3.12

the dominance coefficient of mutations was assumed to be related to the selection coefficient by using an exponential function
as proposed by Caballero and Keightley (1994)
the dominance coefficient of a mutation is taken from a uniform distribution between zero and exp(− ks)
where k is a constant allowing the mean dominance coefficient to be the desired one

for the Mukai model the average coefficient of dominance used was 0.36
a value widely assumed for mild viability mutations
for the new model, a value of 0.20 was used
this has been inferred from a reanalysis of the MA experiment of Ohnishi (1977; see García-Dorado and Caballero 2000)

free recombination among loci was assumed in the production of gametes from each parent
a model with restricted recombination (a total genome length of 1.25 morgans) was also simulated but did not have any substantial effect on the results (not shown)

the distribution of deleterious effects expressed by the later stages had more variance than at earlier stages
a possible explanation is reinforcing epistasis reducing viability in lines carrying several deleterious mutations