SimParameters.R

# Simulation parameters

# ---- Global parameters to simulate the founders genome ----

NFounders <- 5000
nChr <- 30
nSNP <- 1400
nQTL <- 300
nSplit <- 20000

# ---- Genetic parameters for 2 x 2 polygenic traits ----

# The traits are:
#   - BodyWeight_local
#   - TickCountIncidence_local
#   - BodyWeight_exotic
#   - TickCountIncidence_exotic

# Additive variation first
h2 <- c(0.3, 0.1, 0.3, 0.1)
PhenoMean <- c(325, -1.0, 450, -1.5)
PhenoVar <- c(1300, 0.2, 625, 0.2)
AdditVar <- PhenoVar*(h2)

# ---- Correlations ----

# Genetic correlation (rg) between BodyWeight and TickCountIncidence in the same breed.
# Different rg were tested.
CorA <- -0.4
# CorA <- -0.2
# CorA <-  0.0
# CorA <-  0.2
# CorA <-  0.4

# Genetic correlation between the same trait (e.g. BodyWeight) in the two breeds
# (as environmental correlation).
# Different GxE were tested
CorE <- 0.4 # (Strong GxE effects)
# CorE <- 0.6 # (Moderate GxE effects)
# CorE <- 0.8 # (Limited GxE effects)

TraitCor <- matrix(c(1,         CorA,      CorE,      CorA*CorE,
                     CorA,      1,         CorA*CorE, CorE,
                     CorE,      CorA*CorE, 1,         CorA,
                     CorA*CorE, CorE,      CorA,      1),
                   nrow = 4, ncol = 4)

# ---- Dominance ----

# The expected proportion of dominance variance (VD) to phenotype variance (VP) is:
# 0.1 for Body weight and 0.04 for TickCountIncidence
DomMeanDD <- c(0.18, 0.08, 0.18, 0.08)
DomVarDD  <- c(1.0, 1.3, 1.0, 1.3)

# ---- Parameters  to simulate villages and farms ----

nVillages <-10
nFarms <- 200
nMinCows_f <- 8 # Minimum number of cows per farm
nMaxCows_f <- 40 # Maximum number of cows per farm
nBull_v <- 5 # Number of selected bulls per village. Artificial insemination is assumed.
nFarms_v <- nFarms/nVillages

# Trait index for the selection of crosbred animals
TraitIndex <- c(1, 80, 0, 0)   # corresponding to 50% weight for each trait. The weights were estimated based on the phenotypic variance of the two traits
#TraitIndex <- c(1, 30, 0, 0)  # corresponding to 30% weight for tick count and 70 % weight for body weight
#TraitIndex <- c(1, 10, 0, 0)  # corresponding to 10% weight for tick count and 90 % weight for body weight