Preprint: linkage disequilibrium moderates how an evolving genetic background shapes variant-level trait architecture

A preprint deposited on bioRxiv presents a population-genetic model that explicitly incorporates linkage disequilibrium (LD) — the non-random association between alleles at nearby loci — into the description of how a variant's effect on a complex trait is shaped by the genetic background in which it resides.

Most widely used single-variant theoretical frameworks in statistical genetics either omit the coupling between a focal variant and the surrounding haplotype background, or hold the background fixed. The authors argue that this omission introduces systematic biases, particularly when LD is strong, because the background itself evolves under selection and drift in ways that feed back on the apparent effect size and frequency of the focal variant.

The model developed in this work incorporates stabilising selection, recombination, and genetic drift, and derives predictions for variant-level effect-size distributions and allele frequencies that differ materially from standard single-variant theory when LD is substantial. The authors suggest their framework could help explain discrepancies between the expected and observed genetic architecture of complex traits — for example, patterns of effect-size variation across the allele-frequency spectrum.

This work is technical and primarily of interest to population geneticists and statistical geneticists developing or evaluating methods for GWAS interpretation. It has not yet been peer-reviewed.