Local fitness and epistatic effects lead to distinct patterns of linkage disequilibrium in protein-coding genes

Abstract

Selected mutations interfere and interact with evolutionary processes at nearby loci, distorting allele frequency trajectories and correlations between pairs of mutations. A number of recent studies have used patterns of linkage disequilibrium (LD) between selected variants to test for selective interference and epistatic interactions, with some disagreement over interpreting observations from data. Interpretation is hindered by the relative lack of analytic or even numerical expectations for patterns of variation between pairs of loci under the combined effects of selection, dominance, epistasis, and demography. Here, I develop a numerical approach to compute the expected two-locus sampling distribution under diploid selection with arbitrary epistasis and dominance, recombination, and variable population size. I use this to explore how epistasis and dominance affect expected signed LD, including for non-steady-state demography relevant to human populations. Finally, I use whole-genome sequencing data from humans to assess how well we can differentiate modes of selective interactions in practice. I find that positive LD between missense mutations within genes is driven by strong positive allele-frequency correlations between pairs of mutations that fall within the same conserved domain, pointing to compensatory mutations or antagonistic epistasis as the prevailing mode of interaction within but not outside of conserved genic elements. The heterogeneous landscape of both mutational fitness effects and selective interactions within protein-coding genes calls for more refined inferences of the joint distribution of fitness and interactive effects, and the methods presented here should prove useful in that pursuit.

Negative linkage disequilibrium between amino acid changing variants reveals interference among deleterious mutations in the human genome

“…Thus, LD patterns may offer a strategy to infer the relationship between the dominance coefficient and the DFE in outcrossing populations [58]. In order to accurately predict the expected LD decay curve under a given scenario, extensive forward simulations, or recently developed numerical approaches [59], are needed to determine the predicted LD decay as a function of these genomic and selective parameters.…”

“…Future work could quantify the prevalence of epistasis among linked deleterious variants while jointly quantifying the fitness effects of pairs of deleterious variants [59]. Additionally, although it has been shown that two locus statistics such as D' can be used in the detection of epistatic interactions and interference [17,50], it is possible that a combination of summary statistics can better discern between the two.…”

A numerical framework for genetic hitchhiking in populations of variable size

“…As patterns of LD are integral to a variety of existing population genetic tools, the framework described here provides a means of elucidating the impact of selection and demographic history on LD to allow for a better understanding of the patterns observed in genomic data. For example, in recent work by Ragsdale (2021), this framework has been applied to explore the impact of epistasis and dominance on shaping patters of signed LD observed in genic regions. As demonstrated in Section 3.1.1, our novel moment approximation technique is more accurate than the Jackknife approach employed by Ragsdale (2021).…”

“…For example, in recent work by Ragsdale (2021), this framework has been applied to explore the impact of epistasis and dominance on shaping patters of signed LD observed in genic regions. As demonstrated in Section 3.1.1, our novel moment approximation technique is more accurate than the Jackknife approach employed by Ragsdale (2021). Thus, using this novel approximation technique in the analysis performed in Ragsdale (2021) and similar analyses has the potential to increase accuracy and substantiate the conclusions.…”

“…Here, we give an alternative derivation which relies on Dyukin’s formula. Moreover, Ragsdale and Gravel (2019) focus on the neutral case in their study, but the approach has recently been used by Ragsdale (2021) to explore how epistasis and dominance shape expected patterns of signed linkage disequilibrium in protein coding variants. In this work, however, we explore the dynamics of one selected locus and its effect on linked neutral variation in populations with complex demographic histories.…”