sim_trait_env: Simulate a complex trait

In OchoaLab/simgenphen: Simulate Genotypes and Phenotypes

Simulate a complex trait

Description

Mostly a wrapper around simtrait::sim_trait(), but can optionally add a structured environment effect. However, in that case the heritability is no longer as specified.

Usage

sim_trait_env(
  X,
  p_anc,
  m_causal,
  herit,
  env = NA,
  env_var = 1,
  k_subpops = NA,
  fes = FALSE,
  verbose = TRUE
)

Arguments

X	The genotype matrix.
p_anc	The vector of ancestral allele frequencies.
m_causal	The number of causal loci for the trait, selected randomly from among the simulated loci.
herit	The trait heritability.
env	A string describing environment model. Only NA (no environment) or "gcta" are accepted.
env_var	The variance of the environment effect. Ignored if env is NA.
k_subpops	The number of intermediate subpopulations for admixture model. Used for env = 'gcat' only.
fes	If FALSE (default) constructs trait from Random Coefficients (RC) model. If TRUE, the Fixed Effect Sizes (FES) model is used instead.
verbose	If TRUE reports progress, otherwise it is silent.

Value

A list containing the following elements:

• trait: The simulated trait vector.

• causal_indexes: The vector of randomly selected causal loci indexes.

• causal_coeffs: The vector of simulated regression coefficients for causal loci.

Examples

# draw population parameters using `sim_pop` first
# a small example
k_subpops <- 3
data_pop <- sim_pop( n_ind = 50, G = 3, k_subpops = k_subpops )

# then draw genotypes
m_loci <- 100
data_geno <- sim_geno(
    data_pop$admix_proportions_1,
    data_pop$inbr_subpops,
    data_pop$fam,
    data_pop$ids
)

# now draw trait!
m_causal <- 5
herit <- 0.5
data_trait <- sim_trait_env(
    data_geno$X,
    data_geno$p_anc,
    m_causal,
    herit,
    env = 'gcat',
    k_subpops = k_subpops
)

# the simulated trait
data_trait$trait
# indexes of randomly-selected causal loci
data_trait$causal_indexes
# coefficients of causal loci
data_trait$causal_coeffs

OchoaLab/simgenphen documentation built on Feb. 10, 2023, 3:32 p.m.