deleterious mutation - senecio7の日記

Rodríguez-Ramilo ST, Pérez-Figueroa A, Fernández B, Fernández J & Caballero A 2004 Mutation-selection balance accounting for genetic variation for viability in Drosophila melanogaster as deduced from an inbreeding and artificial selection experiment. J Evol Biol 17:528-541.

mutational models assuming a low rate of mutations of large average effect and highly recessive gene action, and others assuming a high rate of mutations of small average effect and close to additive gene action, are compatible with all the observations

classical studies suggested that deleterious mutations occur at a rate in the order of λ ≈ 0.25–0.5 mutations per haploid genome and generation
with average homozygous effect of about s ≈ 0.02
the average coefficient of dominance of minor deleterious mutations (h, where sh is the mutational heterozygous effect) estimated by Ohnishi (1977b) was 0.49
estimates by Mukai and co-workers ranged from around zero, implying almost complete recessive gene action, to around 0.4, implying close to additive gene action, depending on the analyses
the estimates close to zero were generally disregarded and an estimate of h ≈ 0.4 has been generally cited for the average coefficient of dominance of newly arisen minor deleterious mutations

more recent mutation-accumulation studies and reanalyses of previous results (García-Dorado et al., 1999, 2004; García-Dorado & Caballero, 2000), suggest much smaller mutation rates (λ ≈ 0.015)
substantially larger homozygous effects (s ≈ 0.2)
gene action closer to complete recessivity (h ≈ 0.15)
these parameters refer to mutations of minor or moderate effect

it is experimentally difficult to quantify the properties of mutations of extremely low effect (s < 5 × 10^{− 4}; García-Dorado et al., 2004)
it is possible that many of those mutations have passed undetected and were not considered in the above estimates (Keightley & Eyre-Walker, 1999)

the classical model involving many mutations of small effect implies a larger equilibrium additive variance than the more recent model of few mutations of large effect
as well as a larger inbreeding depression and a larger rate of fitness decline due to deleterious fixation in small populations
several earlier studies have focused on the last of these predictions, showing that the observed fitness decline in small populations is incompatible with the classical model but compatible with the more recent model
here, we focus on the first two predictions relative to the additive genetic variance and the inbreeding depression

we carried out simulations resembling the experimental conditions and using different mutational parameters
we assumed a model of variable deleterious mutations in which the genotypic fitnesses at a given locus are 1, 1 − sh and 1 − s for the wild homozygote, the heterozygote and the mutant homozygote
effects are multiplicative among loci
mutations were assumed to occur at a rate λ per haploid genome and generation
effects were sampled from a gamma distribution with shape parameter β and mean effect s
a range of five models was used in the simulations
the first models (λ = 0.25–0.5, s = 0.02–0.03) assume many mutations of small average effect
the last one (λ = 0.015, s = 0.2) assumes few mutations of large average effect
these are the two contrasting models proposed in the literature for spontaneous mutations affecting viability in Drosophila
the remaining models represent intermediate situations
the shape parameter of the gamma distribution (β = 0.2, 0.238, 0.263, 0.714, 1.0, respectively, for the five models) was chosen such that the mutational variance explained by any model is 0.0006
a value that has been experimentally observed (García-Dorado et al., 1999)

no linkage is considered in the simulations
the inclusion of linkage in the simulations appears to be irrelevant
we made simulations assuming a single chromosome of length 1.25 M (the assumed genome length in D. melanogaster)
the results were almost identical to those given in Table 1 for free recombination

we assume a multiplicative model of gene effects among mutations
synergistic epistatic effects are likely to produce a larger inbreeding depression than expected without epistasis, but little effect on the expected additive variance is predicted (Charlesworth & Hughes, 1999)
synergistic epistatic effects of deleterious mutations is supported only by very scarce data, in particular, a single point in the experiment of Mukai (1969a) and another one in the experiment of Whitlock & Bourguet (2000)