R語(yǔ)言怎么把數(shù)據(jù)框每列分割成兩列，然后合并成新的數(shù)據(jù)框？

比如數(shù)據(jù)如下：

kk <- matrix(c("CG","CC","GG","GG","CG","CG","CC","CG","CG","CC","GG","GG"),3,4)

kk <- as.data.frame(kk)

把每一列的字符串分割成兩個(gè)字符，然后各自成為新的列，再進(jìn)行合并。

library(stringr)

split <- function(temp){

??return(str_split_fixed(temp, "", 2))

}

jj <- matrix(0,nrow(kk), ncol(kk))

for(i in 1:ncol(kk)){

??temp <- split(kk[,i])

??jj <- cbind(jj, temp)

}

jj <- jj[,-c(1:4)]

矩陣cbind循環(huán)越往后越慢，而且基因型數(shù)據(jù)通常又很大。

用list存儲(chǔ)，再解成matrix，速度就快得多，試一下。

nrows <- 2000

kk <- matrix(rep(c("CG","CC","GG","GG","CG","CG","CC","CG","CG","CC","GG","GG"), 10000),nrows, byrow=T) %>% 

??as.data.frame

jj <- list()

for(i in 1:ncol(kk)){

??jj[[i]] <- kk[,i] %>% str_split_fixed("", 2)

}

jj <- jj %>% unlist %>% 

??matrix(nrow=nrows, byrow=T) %>% 

??data.frame(stringsAsFactors=F)