/genetic_inheritance

Exploring different types of genetic inheritance: additivity, dominance and epistasis

Primary LanguageR

Mechanisms of genetic inheritance

Exploring different types of genetic inheritance: additivity, dominance and epistasis

This repository hosts code, workflows and ideas on the use of simulated and real phenotypic and genetic data for the study of genetic inheritance. Different scenarios are simulated, considering additive. dominant and epistatic phenotypes. Genetic data are then used to predict and/or approximate the simulated phenotypes by accounting for additivity, dominance and espistasis.

Operational steps

  1. Use data from Nazzicari & Biscarini 2022 (~1,000 samples) to simulate phenotypes under scenarios of extreme epistasis, and add these to the existing fully additive and fully dominant simulated phenotypes
  2. Use data from the $1,000$ genomes project to explore the same scenario on a different population (still diploid genomes, but human)
  3. Use synthetic data from HAPNEST (~1 million samples) to explore the effect of data scale on the results

Human SNP data are likely data from whole-genome sequencing: for the purpose of this work, SNP-array-like data will be extracted, based on the Infinium Global Screening Array with $654,027$ SNP variants (genetic maps and SNP variant lists can be found here).

Modelling additivity, dominance and epistasis

In this work, we are going to follow the model for additivity, dominance and epistasis outlined by Trudy MacKay in: Mackay, T.F., 2014. Epistasis and quantitative traits: using model organisms to study gene–gene interactions. Nature Reviews Genetics, 15(1), pp.22-33.

The scenarios we aim to simulate are those described in Figure 1 (c) of the MacKay's article:

  1. additivty + dominance deviations at one or more loci
  2. additivity + divergent epistatic interaction (locus1_BB has divergent phenotypes (verylow / very-high) depending on the genotype at locus 2 (AA/BB)
  3. additivity + opposite epistatic interaction: the effects of the AA and BB genotypes at locus 1 are inverted depending on the genotypes at locus 2 (AA / BB)

Various degrees of the above three scenarios may be simulated, more or less extreme.

Fixed parameters:

  • MAF: filter at 5% or 10% to retain only SNPs at intermediate frequencies $\rightarrow$ maximise the effect of epistasis (Mackay, 2013)
  • heritability ($h^2$=0.7): we want to have a sizable genetic contribution in order to be able to study the mechanisms of genetic inheritance

Variable parameters:

  • n. QTL: oligogenic vs polygenic traits: e.g. 50 vs 500 vs 5000?
  • scenarios of additivity, dominance and epistasis