Selective mutation accumulation: A computational model of the paternal age effect

Abstract

Motivation: As the mean age of parenthood grows, the effect of parental age on genetic disease and child health becomes ever more important. A number of autosomal dominant disorders show a dramatic paternal age effect due to selfish mutations: substitutions that grant spermatogonial stem cells (SSCs) a selective advantage in the testes of the father, but have a deleterious effect in offspring. In this paper we present a computational technique to model the SSC niche in order to examine the phenomenon and draw conclusions across different genes and disorders. Results: We used a Markov chain to model the probabilities of mutation and positive selection with cell divisions. The model was fitted to available data on disease incidence and also mutation assays of sperm donors. Strength of selective advantage is presented for a range of disorders including Apert's syndrome and achondroplasia. Incidence of the diseases was predicted closely for most disorders and was heavily influenced by the site-specific mutation rate and the number of mutable alleles. The model also successfully predicted a stronger selective advantage for more strongly activating gain-of-function mutations within the same gene. Both positive selection and the rate of copy-error mutations are important in adequately explaining the paternal age effect.