单细胞数据分析计算每个细胞MT-比例

 

1、awk + >>追加实现

for i in $(seq 2 $(head -n 1 a.txt | awk '{print NF}')); do cut -f 1,$i a.txt | awk 'BEGIN{sum =0 } {if(NR != 1) sum += $2} {if($1 ~ /^MT-/ && NR != 1) {sum1 += $2}} END {print 100 * sum1/sum}' >> result.txt; done

 

 

 

2、awk + shell数组实现

sed 1d a.txt | awk '{for(i = 2; i <= NF; i++) {array1[i] += $i; if($1 ~ /^MT-/) {array2[i] += $i}}} END {for(i = 2; i <= NF; i++) {print 100 * array2[i]/array1[i]}}' > result.txt

 

3、R实现

001、

dir()
dat <- read.table("a.txt", header = T)
dim(dat)

mt <- dat[grep("^MT-", rownames(dat)),]
dim(mt)
result <- apply(mt, 2, sum)/apply(dat, 2, sum) * 100
head(result)
tail(result)

 

002、

dir()
dat <- read.table("a.txt", header = T)

sum1 <- rep(0, ncol(dat))
sum2 <- rep(0, ncol(dat))

for (i in 1:ncol(dat)) {
 for (j in 1:nrow(dat)) {
   sum1[i] <- sum1[i] + dat[j,i]
   if (substring(rownames(dat)[j],1, 3) == "MT-") {
     sum2[i] <- sum2[i] + dat[j, i]
   }
 }
}
head(sum1)
head(sum2)
result <- sum2/sum1
head(result)

 

003、

dir()
dat <- read.table("a.txt", header = T)

sum1 <- rep(0, ncol(dat))
sum2 <- rep(0, ncol(dat))
mt <- dat[grep("^MT-", rownames(dat)),]

for (i in 1:ncol(dat)) {
  sum1[i] <- sum(dat[,i])
  sum2[i] <- sum(mt[,i])
}
result <- sum2/sum1
head(result)

 

004、

dat <- read.table("a.txt", row.names = 1, header = T)
dim(dat)
result <- apply(dat[grepl("^MT-",rownames(dat)),], 2, sum)/apply(dat, 2, sum) * 100
head(result)