GWA.R.txt

# load libraries
chooseCRANmirror(ind=73)
# Load packages
Pckg.Lst <-c("vroom","readxl","tidyverse", "bigmemory", "biganalytics", "parallel", "MASS")
      package.check <- lapply(
        Pckg.Lst,
        FUN = function(x) {
        if (!require(x, character.only = TRUE)) {
          install.packages(x, dependencies = TRUE)
        library(x, character.only = TRUE)}})

# load gapit functions to convert to numeric format.
source("http://zzlab.net/GAPIT/GAPIT.library.R")
source("http://zzlab.net/GAPIT/gapit_functions.txt")
        # I made updates so the writing the output takes less time
source("../Input.dat/Bill.Truncated.GAPIT.Res.txt")

# Combine all numeric input genotypic data
C1.dat <- vroom("Chr01/GAPIT.Genotype.Numerical.txt", col_names=T)
C1.map <- vroom("Chr01/GAPIT.Genotype.map.txt", col_names=T)
C1.map$Chromosome <- 1
C2.dat <- vroom("Chr02/GAPIT.Genotype.Numerical.txt", col_names=T)
C2.map <- vroom("Chr02/GAPIT.Genotype.map.txt", col_names=T)
C2.map$Chromosome <- 2
C3.dat <- vroom("Chr03/GAPIT.Genotype.Numerical.txt", col_names=T)
C3.map <- vroom("Chr03/GAPIT.Genotype.map.txt", col_names=T)
C3.map$Chromosome <- 3
C4.dat <- vroom("Chr04/GAPIT.Genotype.Numerical.txt", col_names=T)
C4.map <- vroom("Chr04/GAPIT.Genotype.map.txt", col_names=T)
C4.map$Chromosome <- 4
C5.dat <- vroom("Chr05/GAPIT.Genotype.Numerical.txt", col_names=T)
C5.map <- vroom("Chr05/GAPIT.Genotype.map.txt", col_names=T)
C5.map$Chromosome <- 5
C6.dat <- vroom("Chr06/GAPIT.Genotype.Numerical.txt", col_names=T)
C6.map <- vroom("Chr06/GAPIT.Genotype.map.txt", col_names=T)
C6.map$Chromosome <- 6
C7.dat <- vroom("Chr07/GAPIT.Genotype.Numerical.txt", col_names=T)
C7.map <- vroom("Chr07/GAPIT.Genotype.map.txt", col_names=T)
C7.map$Chromosome <- 7
C8.dat <- vroom("Chr08/GAPIT.Genotype.Numerical.txt", col_names=T)
C8.map <- vroom("Chr08/GAPIT.Genotype.map.txt", col_names=T)
C8.map$Chromosome <- 8
C9.dat <- vroom("Chr09/GAPIT.Genotype.Numerical.txt", col_names=T)
C9.map <- vroom("GAPIT.Num.Con/Chr09/GAPIT.Genotype.map.txt", col_names=T)
C9.map$Chromosome <- 9
mast.map <- rbind(C1.map, C2.map, C3.map, C4.map, C5.map, C6.map, C7.map, C8.map, C9.map)
C2.dat <- C2.dat[,-1]
C3.dat <- C3.dat[,-1]
C4.dat <- C4.dat[,-1]
C5.dat <- C5.dat[,-1]
C6.dat <- C6.dat[,-1]
C7.dat <- C7.dat[,-1]
C8.dat <- C8.dat[,-1]
C9.dat <- C9.dat[,-1]
mast.dat <- cbind(C1.dat, C2.dat, C3.dat, C4.dat, C5.dat, C6.dat, C7.dat, C8.dat, C9.dat)

#load kinship file calcualted while the file was being converted to numeric
myKI <- read.csv("GAPIT.Kin.VanRaden.csv", head=F)

#load phenotype file
pheno <- read_csv("GWA.Phenotype.csv", col_names=T)
pheno <- pheno[,c(1,4,9,10] 
#convert genotype file to dataframe from tibble
mast.dat <- as.data.frame(mast.dat)
# convert phenotype to dataframe
pheno <- as.data.frame(pheno)
# convert kinship file to dataframe
myKI <- as.data.frame(myKI)

# read PCA
CV.pop <- read.csv("GAPIT.PCA.csv", head=T)

# run GAPIT model
myGAPIT <- GAPIT(
  Y=pheno,
  GD=mast.dat,
  GM=mast.map,
  KI=myKI,
  CV=CV.pop,
  SNP.MAF = 0.05, # reset this threshold. Phenotype filtering, and resulting samples made initial MAF 0.05 less effective. Reset in GAPIT.
  model="MLM")