Importing Real Data"
In AlphaSimR: Breeding Program Simulations

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(AlphaSimR)

AlphaSimR provides several functions that allow users to create simulatins using real genotypic data.

# Step 1 ----
# Load in founder haplotypes.
# This requires a genetic map and phased genotypes

# Load the genetic map in the below format
# Format: Marker name, Chromosome, Position (in Morgans) 
# Modeling 10 chromosomes with 10 loci each
# Loci are equally space along 1 Morgan chromosomes
genMap = data.frame(marker = paste0("x", 1:100),
                    chromosome = rep(1:10, each = 10),
                    position = rep(seq(from = 0,
                                       to = 1,
                                       length.out = 10),
                                   times = 10))

# Load the haplotypes in the below format
# Modeling 10 individuals
# There will need to be 20 rows (# of individuals times 2, since each has 2 haplotypes)
# Each individual will have 100 loci based on the map above
# Generating just random 0s and 1s
haplo = matrix(sample(0:1, size=2*10*100, replace=TRUE), nrow=2*10, ncol=100)

# Assign marker names to the haplotypes
# These can be in any order, because the software will order them based 
# on the genetic map automatically
colnames(haplo) = genMap$marker

# Load a pedigree in the below format (optional)
# Pedigree will be for just the 10 individuals represented in the haplotypes
ped = data.frame(id = letters[1:10],
                 mother = rep(0, 10),
                 father = rep(0, 10))

# Create the founder population
# Uses the data structures loaded above
founderPop = importHaplo(haplo = haplo,
                         genMap = genMap,
                         ploidy = 2,
                         ped = ped)

# Step 2 ----
# Set simulation parameters
# Initialize parameters with founder haplotypes
SP = SimParam$new(founderPop)

# Load your own QTL effects (optional)
# This is useful if you want to model known QTL, estimated effects from GS, or
# you just want to create your own distribution for effects.
# The marker names must match the names given in the map above.
# You can model additive effects, dominance effects, and an intercept.
qtlEffects = data.frame(marker = c("x1", "x11"),
                        aditiveEffect = c(1, -1))

# Import in SimParam
SP$importTrait(markerNames = qtlEffects$marker,
               addEff = qtlEffects$aditiveEffect,
               name = "Your_Trait")

# Step 3 ----
# Create a population from the founder haplotypes
pop = newPop(founderPop)

# The population now works like any other AlphaSimR population
gv(pop)
getPed(pop)