Question

I have two columns in a data frame, and I have been able to delete all duplicate rows with unique( ) – Works a treat.

But now I want to delete rows were the values are the same, irrespective of which column they are in. like…

data1    data2
data3    data2
data2    data1
data2    data3

Should be simplified to

data1    data2
data3    data2

since rows 3 and 4 are the same as 1 and 2.

Any ideas?

Answer 1

First sort each row column-wise (using apply and sort), then use unique:

dat <- read.table(text="
data1    data2
data3    data2
data2    data1
data2    data3")

unique(t(apply(dat, 1, sort)))
     [,1]    [,2]   
[1,] "data1" "data2"
[2,] "data2" "data3"