A quantitative genetic model for indirect genetic effects and genomic imprinting under random and assortative mating

Ilse Krätschmer, Matthew R. RobinsonGenetics 2026 · DOI

TL;DR

This paper introduces a novel quantitative genetic model that simultaneously accounts for direct genetic effects, indirect genetic effects from parents and siblings, and genomic imprinting. The model reveals that assortative mating (AM) significantly complicates the interpretation of genetic variance by inducing covariances between these effects, and that both indirect genetic effects and AM are crucial for explaining patterns of relatedness in human traits like height.

What Problem Does This Paper Address?

Understanding the genetic basis of complex traits is a long-standing challenge. Traditionally, quantitative genetics models have focused on direct genetic effects (an individual's own DNA influencing their phenotype). However, it's increasingly recognized that parental genotypes can influence offspring phenotypes indirectly, through environmental mediation (e.g., maternal care, upbringing) or even through epigenetic modifications. Furthermore, genomic imprinting, where an allele's expression depends on its parental origin, adds another layer of complexity. Existing models often struggle to disentangle these various influences, especially when individuals choose mates with similar traits (assortative mating, AM), which further confounds genetic associations. This paper aims to develop a unified theoretical framework to simultaneously model and quantify these diverse genetic influences and their interactions under both random mating (RM) and assortative mating.

Methods

The authors develop a quantitative genetic model that extends previous work by incorporating direct, indirect maternal, indirect paternal, and parent-of-origin (imprinting) genetic effects. They derive the population genetic variance for a single locus under random mating and then extend this to include indirect sibling effects. A key contribution is the extension of the model to account for assortative mating (AM), which alters genotype frequencies and introduces correlations between alleles within and across loci. They derive the genetic variance under AM, highlighting how it increases due to increased homozygosity and correlations. The model is further extended to a multilocus framework, considering correlations between loci. Finally, they validate their theoretical predictions through forward-in-time simulations using realistic linkage disequilibrium data from the 1000 Genomes Project and compare their model's predictions to real-world data on human height from the UK Biobank.

Datasets

  • 1000 Genomes Project: Used to generate realistic linkage disequilibrium (LD) data for forward-in-time simulations.
  • UK Biobank: Used to obtain estimates for heritability and assortative mating for human height, against which the model's predictions were compared.

Key Insights and Ideas

  • Confounding is the Norm: The model demonstrates that direct, indirect parental, and parent-of-origin effects are inherently confounded, making it difficult to attribute phenotypic variance to a single source without a comprehensive model.
  • Assortative Mating Amplifies Complexity: AM doesn't just increase homozygosity; it induces covariances between different types of genetic effects (direct, indirect, imprinting), significantly altering the genetic variance and making interpretation even more challenging.
  • Shared Parental Effects: AM can lead to parents with similar indirect genetic effects on their children, creating a "shared parental effect" that could explain cross-trait AM patterns observed in humans. This covariance of indirect genetic effects is a novel aspect highlighted by the model.
  • Sibling Differences Offer a Unique Window: Analyzing phenotypic differences between siblings can help disentangle indirect parental effects (which cancel out in differences) from direct and parent-of-origin effects under RM. However, under AM, sibling differences are also affected by inter-locus correlations.
  • Indirect Effects and AM are Key for Height: Explaining the observed patterns of phenotypic correlation with genomic relationship for human height requires the inclusion of both indirect genetic effects and assortative mating. Neither alone is sufficient.

Key Results

  • The theoretical model accurately predicts variance components in simulations across various scenarios of direct, indirect, and imprinting effects, both under random mating (RM) and assortative mating (AM).
  • Under RM, the inclusion of indirect genetic and imprinting effects increases population-level variance.
  • Under AM, the genetic variance is significantly altered due to induced correlations between alleles within and across loci. For sibling differences, AM reduces variance more substantially than it increases variance in trios.
  • When applied to human height data, the model predicts that both indirect genetic effects and AM are necessary to capture the observed phenotypic correlations across different degrees of genomic relationship. The model's predictions for close relatives under AM and indirect effects align better with UK Biobank data than RM alone.

Limitations and Future Work

  • Simplifications: The model currently ignores dominance, genotype-environment interaction, and epistasis. Future work could incorporate these factors.
  • Mate Choice Modeling: The current AM model uses a simplified approach. Future research could explore more complex mating strategies and their impact on genetic variance.
  • Data Requirements: The authors emphasize that resolving the precise contributions of direct, indirect, and epigenetic effects to complex traits will require large-scale family data.
  • Distinguishing Imprinting and Indirect Effects: The paper notes the inherent difficulty in distinguishing between genomic imprinting and indirect genetic effects, as they can produce similar patterns. Further research is needed to disentangle these phenomena.
  • Causal Interpretation: The authors acknowledge that even with sophisticated models, interpreting causal relationships from genetic data can be challenging due to the high correlation between markers and effects, especially in the context of AM.

Summary updated

Abstract

An individual's phenotype reflects a complex interplay of the direct effects of their DNA, epigenetic modifications of their DNA induced by their parents, and indirect effects of their parents' DNA. Here, we derive how the genetic variance within a population is changed under the influence of indirect maternal, paternal and parent-of-origin effects under random mating. We also consider indirect effects of a sibling, in particular how the genetic variance is altered when looking at the phenotypic difference between two siblings. The calculations are then extended to include assortative mating (AM), which alters the variance by inducing increased homozygosity and correlations within and across loci. AM likely leads to covariance of parental genetic effects, a measure of the similarity of parents in the indirect effects they have on their children. We propose that this assortment for parental characteristics, where biological parents create similar environments for their children, can create shared parental effects across traits and the appearance of cross-trait AM. Our theory shows how the resemblance among relatives increases under both AM, indirect and parent-of-origin effects. When our model is used to predict correlations among relatives in human height, we find that explaining the patterns observed in real data requires both indirect genetic effects and assortative mating. The degree to which direct, indirect and epigenetic effects shape the phenotypic variance of complex traits remains an open question that requires large-scale family data to be resolved.
BibTeX
@inproceedings{kratschmer2026quantitative,
  author = {Ilse Krätschmer and Matthew R. Robinson},
  title = {{A quantitative genetic model for indirect genetic effects and genomic imprinting under random and assortative mating}},
  booktitle = {Genetics},
  year = {2026},
  doi = {10.1093/genetics/iyag042},
}
Figure