BMTME (version 1.0.4)

BMTME: Bayes Multi-Trait Milti-Environment Model (BMTME)

Description

The Bayesian Multi-Trait Multi-Environment models (BMTME) package was developed to implement...

Usage

BMTME(Y, X, Z1, Z2, nIter = 1000L, burnIn = 300L, thin = 2L,
  bs = ceiling(dim(Z1)[2]/6), parallelCores = 1, digits = 4,
  progressBar = TRUE, testingSet = NULL)

Arguments

Y

(matrix) Phenotypic response where each column is a different trait.

X

(matrix) Matrix design for the environment effects.

Z1

(matrix) Matrix design for the genetic effects.

Z2

(matrix) Matrix design for the genetic effects interaction with the environment effects.

nIter

(integer) Number of iterations to fit the model.

burnIn

(integer) Number of items to burn at the beginning of the model.

thin

(integer) Number of items to thin the model.

bs

(integer) Number of groups.

parallelCores

(integer) Number of cores to use.

digits

(integer) Number of digits of accuracy in the results.

progressBar

(Logical) Show the progress bar.

testingSet

(object or vector) Crossvalidation object or vector with the positions to use like testing in a cross-validation test.

References

Montesinos-Lopez, O.A., Montesinos-Lopez, A., Crossa, J., Toledo, F.H., Perez-Hernandez, O., Eskridge, K.M., <U+2026> Rutkoski, J. (2016). A Genomic Bayesian Multi-trait and Multi-environment Model. G3: Genes|Genomes|Genetics, 6(9), 2725<U+2013>2744. https://doi.org/10.1534/g3.116.032359.

Examples

# NOT RUN {
  data("WheatIranianToy")

  # Matrix Design
  LG <- cholesky(genoIranianToy)
  ZG <- model.matrix(~0 + as.factor(phenoIranianToy$GID))
  Z.G <- ZG %*% LG
  Z.E <- model.matrix(~0 + as.factor(phenoIranianToy$Env))
  ZEG <- model.matrix(~0 + as.factor(phenoIranianToy$GID):as.factor(phenoIranianToy$Env))
  G2 <- kronecker(diag(length(unique(phenoIranianToy$Env))), data.matrix(genoIranianToy))
  LG2 <- cholesky(G2)
  Z.EG <- ZEG %*% LG2

  #Pheno
  Y <- as.matrix(phenoIranianToy[, -c(1, 2)])

  #Check fitting
  fm <- BMTME(Y = Y, X = Z.E, Z1 = Z.G, Z2 = Z.EG,
              nIter = 10000, burnIn = 5000, thin = 2, bs = 50)
  fm

  # Check predictive capacities of the model
  pheno <- data.frame(GID = phenoIranianToy[, 1],
                      Env = phenoIranianToy[, 2],
                      Response = phenoIranianToy[, 3])
  CrossV <- CV.RandomPart(pheno, NPartitions = 4, PTesting = 0.2, set_seed = 123)

  pm <- BMTME(Y = Y, X = Z.E, Z1 = Z.G, Z2 = Z.EG,
              nIter = 10000, burnIn = 5000, thin = 2,
              bs = 50, testingSet = CrossV)
  pm
# }
# NOT RUN {
# }