How to stack multiple columns into one using R

Question

I have the following data frame:

A <- c(3,5,6,7)
B <- c(2,4,5,3)
C <- c(4,6,7,8)
D <- c(2,4,5,3)
gene <- c(1,2,3,4)
df <- data.frame(gene,A,B,C,D)
df
  gene A B C D
1    1 3 2 4 2
2    2 5 4 6 4
3    3 6 5 7 5
4    4 7 3 8 3

How can I stack each lettered column into one new column called "count" such that there is another new column called "sample" that keeps track of the original column from which each count value came frame (ie. I would like the following output):

count sample
  3     A  
  5     A
  6     A
  7     A
  2     B
  4     B
  5     B
  3     B
  4     C
  6     C
  7     C
  8     C
  2     D
  4     D
  5     D
  3     D

Sorry this is difficult to explain but the output data frame above should make it clear.

Thanks

score 3 · Accepted Answer · answered Oct 29 '21 at 19:24

3

In base R, use stack after removing the first column

out <- stack(df[-1])
names(out) <- c("count", "sample")

answered Oct 29 '21 at 19:24

akrun

874,273
37
540
662

score 3 · Answer 2 · answered Oct 29 '21 at 19:26

We could use pivot_longer:

library(tidyr)
library(dplyr)
df %>% 
  pivot_longer(
    cols = -gene,
    names_to = "sample",
    values_to = "count"
  ) %>% 
  select(-gene) %>% 
  arrange(sample)

   sample count
   <chr>  <dbl>
 1 A          3
 2 A          5
 3 A          6
 4 A          7
 5 B          2
 6 B          4
 7 B          5
 8 B          3
 9 C          4
10 C          6
11 C          7
12 C          8
13 D          2
14 D          4
15 D          5
16 D          3

How to stack multiple columns into one using R

2 Answers2