CODEWITHSUNDEEP
rreshape2
Michael Hall
Michael Hall

Reputation: 3330

I want to turn column names into an identifying column in R

Hey guys I'm pretty new to R so apologies if this is a dumb question. I have a data frame that looks like this:

Gelatin Geltrex Combo Control
1   0.761   0.732 0.557   0.723
2   0.704   0.712 0.764   0.540
3   0.580   0.435 0.351   0.537

I want to end up with 2 columns; one with the numbers in the row and another column with the current column names as identifiers for the numbers?

Upvotes: 4

Views: 6587

Answers (2)

Patrick Cree
Patrick Cree

Reputation: 39

Maybe you can use gather from the tidyr package

> library(tidyr)
> df = data.frame(a = c(1,2,3), b = c(4,5,6), d = c(7, 8, 9))
> newdf = gather(df, columnNames, values)

  columnNames values
1           a      1
2           a      2
3           a      3
4           b      4
5           b      5
6           b      6
7           d      7
8           d      8
9           d      9

Upvotes: 1

akrun
akrun

Reputation: 886948

If we need row/column index we can use row and col to get the row/column index and create a new dataset with data.frame. 

df2 <- data.frame(Row= c(row(df1)), 
               Col=names(df1)[col(df1)], stringsAsFactors=FALSE)

Or another option is melt from library(reshape2). We convert the 'data.frame' to 'matrix' and melt to get the row/column names as the first two columns.

library(reshape2)
melt(as.matrix(df1))[-3]

EDIT: If we need the value and column names as the two columns, just melt or use stack as @thelatemail mentioned

melt(df1)

Or stack from base R

stack(df1)

Upvotes: 5

Related Questions