unstr_asr_input: Simulate genetic values based on an unstructured model for GxE interaction - `AlphaSimR` input parameters

Description

Creates a list of input parameters for `AlphaSimR` to simulate genetic values in multiple environments for one or more traits based on an unstructured model for genotype-by-environment (GxE) interaction.
This function utilises the ability of `AlphaSimR` to simulate correlated traits. The wrapper function unstr_asr_input() is used to specify the input parameters required in `AlphaSimR`, and can handle separable and non-separable structures between traits and environments (see below). After simulating the genetic values, the wrapper function unstr_asr_output can be used to generate a data frame with output values.

Usage

unstr_asr_input(
  ntraits = 1,
  nenvs = 2,
  mean = 0,
  var = 1,
  Tvar = NULL,
  Evar = NULL,
  corA = NULL,
  TcorA = NULL,
  EcorA = NULL,
  meanDD = NULL,
  varDD = NULL,
  TvarDD = NULL,
  EvarDD = NULL,
  corDD = NULL,
  TcorDD = NULL,
  EcorDD = NULL,
  relAA = NULL,
  TrelAA = NULL,
  ErelAA = NULL,
  corAA = NULL,
  TcorAA = NULL,
  EcorAA = NULL
)

Value

A list with input parameters for `AlphaSimR`, which are used to simulate correlated genetic values based on an unstructured model for GxE interaction.

Arguments

ntraits: Number of traits to be simulated.
nenvs: Number of environments to be simulated (minimum of two).
mean: A vector of mean genetic values for each environment-within-trait combination. If only one value is specified, all combinations will be assigned the same mean.
var: A vector of genetic variances for each environment-within-trait combination. If only one value is specified, all combinations will be assigned the same variance.
Alternatively, if a separable structure between traits and environments is desired, Tvar and Evar can be specified.
Tvar: A vector of genetic variances for each trait. Must be provided in combination with Evar.
Alternatively, var can be specified.
Evar: A vector of genetic variances for each environment. Must be provided in combination with Tvar.
Alternatively, var can be specified.
corA: A matrix of additive genetic correlations between environment-within-trait combinations. By default, a diagonal matrix is constructed.
Alternatively, TcorA and EcorA can be specified.
TcorA: A matrix of additive genetic correlations between traits. Must be provided in combination with EcorA.
Alternatively, corA can be specified.
EcorA: A matrix of additive genetic correlations between environments. Must be provided in combination with TcorA.
Alternatively, corA can be specified.
meanDD: A vector of mean dominance degrees for each environment-within-trait combination (similar to mean). If only one value is specified, all combinations will be assigned the same mean. By default, meanDD = NULL and dominance is not simulated.
varDD: A vector of dominance degree variances for each environment-within-trait combination (similar to var). If only one value is specified, all combinations will be assigned the same variance.
Alternatively, if a separable structure between traits and environments is desired, TvarDD and EvarDD can be specified.
TvarDD: A vector of dominance degree variances for each trait (similar to Tvar). Must be provided in combination with EvarDD.
Alternatively, varDD can be specified.
EvarDD: A vector of dominance degree variances for each environment (similar to Evar). Must be provided in combination with TvarDD.
Alternatively, varDD can be specified.
corDD: A matrix of dominance degree correlations between environment-within-trait combinations (similar to corA). If not specified and dominance is simulated, a diagonal matrix is constructed.
Alternatively, TcorDD and EcorDD can be specified.
TcorDD: A matrix of dominance degree correlations between traits (similar to TcorA). Must be provided in combination with EcorDD.
Alternatively, corDD can be specified.
EcorDD: A matrix of dominance degree correlations between environments (similar to EcorA). Must be provided in combination with TcorDD.
Alternatively, corDD can be specified.
relAA: A vector defining the relative magnitude of additive-by-additive (epistatic) variance to additive genetic variance for each environment-within-trait combination, that is in a diploid organism with allele frequency of 0.5. If only one value is specified, all environment-within-trait combinations will be assigned the same value. By default, relAA = NULL and epistasis is not simulated.
Alternatively, if a separable structure between traits and environments is desired, TrelAA and ErelAA can be specified.
TrelAA: A vector defining the relative magnitude of epistatic variance to additive genetic variance for each trait. Must be provided in combination with ErelAA.
Alternatively, relAA can be specified.
ErelAA: A vector defining the relative magnitude of epistatic variance to additive genetic variance for each environment. Must be provided in combination with TrelAA.
Alternatively, relAA can be specified.
corAA: A matrix of epistatic correlations between environment-within-trait combinations (similar to corA). If not specified and epistasis is simulated, a diagonal matrix is constructed.
Alternatively, TcorAA and EcorAA can be specified.
TcorAA: A matrix of epistatic correlations between traits (similar to TcorA). Must be provided in combination with EcorAA.
Alternatively, corAA can be specified.
EcorAA: A matrix of epistatic correlations between environments (similar to EcorA). Must be provided in combination with TcorAA.
Alternatively, corAA can be specified.

Details

unstr_asr_input can handle separable and non-separable structures between traits and environments.

For separable structures, provide (1) Tvar & Evar, and (2) TcorA & EcorA.
For non-separable structures, provide (1) var, and (2) corA.

Note: `AlphaSimR` can simulate different biological effects (see: SimParam).

For additive traits use addTraitA().
For additive + dominance traits use addTraitAD().
For additive + epistatic traits use addTraitAE().
For additive + dominance + epistatic traits use addTraitADE().

Check the useVarA argument of these functions when simulating non-additive traits.

Examples

Run this code

# Simulate genetic values with 'AlphaSimR' for two additive + dominance traits
# in two environments based on an unstructured model.

# 1. Define the genetic architecture of the simulated traits.
# Mean genetic values and mean dominance degrees.
mean <- c(4.9, 5.4, 235.2, 228.5) # Trait 1 x 2 environments, Trait 2 x 2 environments
meanDD <- c(0.4, 0.4, 0.1, 0.1) # Trait 1 and 2, same value for both environments

# Additive genetic variances and dominance degree variances.
var <- c(0.086, 0.12, 15.1, 8.5) # Trait 1 x 2 environments, Trait 2 x 2 environments
varDD <- 0.2 # Same value for all environment-within-trait combinations

# Additive genetic correlations between the two simulated traits.
TcorA <- matrix(c(
  1.0, 0.6,
  0.6, 1.0
), ncol = 2)

# Additive genetic correlations between the two simulated environments.
EcorA <- matrix(c(
  1.0, 0.2,
  0.2, 1.0
), ncol = 2)

# Dominance degree correlations between the four environment-within-trait combinations.
corDD <- diag(4) # Assuming independence

input_asr <- unstr_asr_input(
  ntraits = 2,
  nenvs = 2,
  mean = mean,
  var = var,
  TcorA = TcorA,
  EcorA = EcorA,
  meanDD = meanDD,
  varDD = varDD,
  corDD = corDD
)

Run the code above in your browser using DataLab