dataframe - r remove rows from a data frame that contain a duplicate of either combination of 2 columns -

this question has answer here:

• remove duplicate column pairs, sort rows based on 2 columns [duplicate] 3 answers

i trying remove rows data frame contain either combination of 2 columns. example, following code:

vct <- c("a", "b", "c") <- b <- vct combo <- expand.grid(a,b) #generate posible combinations combo <- combo[!combo[,1] == combo[,2],] #removes rows matching column 

generates data frame:

 var1 var2 2    b    3    c    4       b 6    c    b 7       c 8    b    c 

how can remove rows duplicates of combination of 2 columns, i.e. #4 b removed because #2 b present? resulting data frame this:

 var1 var2 2    b    3    c    4    c    b 

we can sort row using apply margin=1, transpose (t) output, use duplicated logical index of duplicate rows, negate (!) rows not duplicated, , subset dataset.

combo[!duplicated(t(apply(combo, 1, sort))),] #   var1 var2 #2    b    #3    c    #6    c    b