Mutation bias reflects natural selection in Arabidopsis thaliana

Abstract

Since the first half of the twentieth century, evolutionary theory has been dominated by the idea that mutations occur randomly with respect to their consequences1. Here we test this assumption with large surveys of de novo mutations in the plant Arabidopsis thaliana. In contrast to expectations, we find that mutations occur less often in functionally constrained regions of the genome—mutation frequency is reduced by half inside gene bodies and by two-thirds in essential genes. With independent genomic mutation datasets, including from the largest Arabidopsis mutation accumulation experiment conducted to date, we demonstrate that epigenomic and physical features explain over 90% of variance in the genome-wide pattern of mutation bias surrounding genes. Observed mutation frequencies around genes in turn accurately predict patterns of genetic polymorphisms in natural Arabidopsis accessions (r = 0.96). That mutation bias is the primary force behind patterns of sequence evolution around genes in natural accessions is supported by analyses of allele frequencies. Finally, we find that genes subject to stronger purifying selection have a lower mutation rate. We conclude that epigenome-associated mutation bias2 reduces the occurrence of deleterious mutations in Arabidopsis, challenging the prevailing paradigm that mutation is a directionless force in evolution.

How Sequence Context-Dependent Mutability Drives Mutation Rate Variation in the Genome

“…We know germline mutation rates have been associated with other genomic features such as histone markers, transcription rate, and replication timing ( Chen et al 2017 ; Supek and Lehner 2019 ). Indeed, our model makes a prediction that mutation rate will be higher in genes, which is in contrast to a recent empirical study in Arabidopsis thaliana that found a lower mutation rate in coding sequence ( Monroe et al 2022 ). Their predictive model of mutation differs from ours in that it depends strongly on empirically estimated epigenetic factors that are not represented in our model.…”

Complete sequence of a 641-kb insertion of mitochondrial DNA in the Arabidopsis thaliana nuclear genome

“…The C➔T transitions that dominate the numt mutation spectrum are a hallmark of the abundant 5-methylcytosine (5mC) modifications at CpG and CHG sites in plant nuclear genomes (Vanyushin and Ashapkin 2011; Weng, et al 2019; Naish, et al 2021; Monroe, et al 2022). We found that 88 of the 235 C➔T observed SNVs occur at CpG sites, and an additional 87 occur at CHG sites.…”

Large haplotypes highlight a complex age structure within the maize pan-genome

“…2, Sup Fig 10). Before doing so, we masked all annotated genes and unmethylated (potential regulatory) regions under the assumption that intergenic sequences would be less likely to be constrained by selection (Lynch et al 2016; Monroe et al 2022). The divergence dating confirmed that cenhaps on seven chromosomes fall into two haplogroups with divergence times ranging from ∼450-130 mya.…”