Abstract
Understanding when, where and which mutations are mostly likely to occur impacts many areas of evolutionary biology, from genetic diseases to phylogenetic reconstruction. Africans and non-African humans differ in the mutability of different triplet base combinations. Africans and non-Africans also differ in mutation rate, possibly because heterozygosity is mutagenic, such that diversity lost when humans expanded out of Africa also lowered the mutation rate. I show that these phenomena are linked: as flanking heterozygosity increases, some triplets become progressively more mutable while others become less so. Africans and non-African show near-identical patterns of dependence on heterozygosity. Thus, the striking differences in triplet mutation frequency between Africans and non-Africans, at least in part, seem to be an emergent property, driven by the way changes in heterozygosity ‘out of Africa’ have differentially impacted the mutability of different triplets. As heterozygosity decreased, the mutation spectrum outside Africa became enriched for triplet mutations that are favoured by low heterozygosity while those favoured by high heterozygosity became relatively rarer.
Highlights
Mutations are classically viewed as occurring more or less randomly, to the extent that they form a molecular clock
G/Cs in triplet different mean age in each of the different populations [25] and that the heterozygosity surrounding a site where and when a mutation occurred in an individual is only crudely approximated by population level heterozygosity measured in current samples. With such similar dependencies of triplet mutability on heterozygosity, the large loss of heterozygosity ‘out of Africa’ must have impacted the mutation spectrum and may well go a long way towards explaining the shifts in mutation probability reported by Harris
The slopes of the relationships are all very similar, even though the intercepts differ. These trends suggest that recent mutations in Africa, being on average in regions with relatively higher heterozygosity, should be skewed towards the types that become more likely in higher heterozygosity regions
Summary
Mutations are classically viewed as occurring more or less randomly, to the extent that they form a molecular clock. Different mean age in each of the different populations [25] and that the heterozygosity surrounding a site where and when a mutation occurred in an individual is only crudely approximated by population level heterozygosity measured in current samples With such similar dependencies of triplet mutability on heterozygosity, the large loss of heterozygosity ‘out of Africa’ must have impacted the mutation spectrum and may well go a long way towards explaining the shifts in mutation probability reported by Harris. The slopes of the relationships are all very similar, even though the intercepts differ Together, these trends suggest that recent mutations in Africa, being on average in regions with relatively higher heterozygosity, should be skewed towards the types that become more likely in higher heterozygosity regions. These comprise low coverage genome sequences for 2504 individuals drawn from 26 modern human populations spread across five main geographical regions: Africa (seven populations), Europe (five populations), Central Southern Asia (five populations), East Asia (five populations) and the Americas (four populations)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
