R语言统计 plink格式数据样本缺失率

 

1、R脚本

dir()
dat <- read.table("outcome.ped")
dat
id <- dat[,2]
dat <- dat[, -(1:6)]

result <- data.frame()

for (i in 1:nrow(dat)) {
  count = 0
  for (j in 1:ncol(dat)) {
    if (dat[i,j] == 0) {
      count = count + 1
    }
  }
  temp <- c(count/2,ncol(dat)/2, count/ncol(dat))
  result <- rbind(result, temp)
}

result <- cbind(id, result)
colnames(result)[2:4] <- c("misssite","totalsite","rate")
result

 

 

2、plink验证

[root@centos79 test]# ls
outcome.map  outcome.ped  result.txt  test.sh
[root@centos79 test]# plink --file outcome --missing --out verify > /dev/null; rm *.log *.nosex
[root@centos79 test]# ls
outcome.map  outcome.ped  result.txt  test.sh  verify.imiss  verify.lmiss
[root@centos79 test]# cat verify.imiss
 FID  IID MISS_PHENO   N_MISS   N_GENO   F_MISS
 DOR    1          Y        4        8      0.5
 DOR    2          Y        4        8      0.5
 DOR    3          Y        3        8    0.375
 DOR    4          Y        2        8     0.25
 DOR    5          Y        1        8    0.125
 DOR    6          Y        1        8    0.125