Home R how to identify duplicate rows in all(multiple columns) and or(multiple columns)?

Questions

R how to identify duplicate rows in all(multiple columns) and or(multiple columns)?

February 10, 2023

I want to identify duplicate rows in a data frame based on two types of conditions:

1: all(multiple columns), all the elements in the multiple columns should be the same.

2: any(multiple columns), if only one of the elements in the multiple columns is the same, then they are considered as replicates.

getRepplicate <- function(df, allCol = "", anyCol = "") {
# both condition 1 and condition 2 are fit, then they are considered as replicated rows
}

For example:

df <- data.frame(
  a = c(1, 1, 2, 3, 4, 1, 1, 3),
  b = c(1, 2, 2, 3, 4, 1, 1, 3), 
  d = c("x", "y", "z", "x", "x", "y", "x", "x"),
  e = c("x", "y", "z", "x", "x", "x", "z", "x")
)
> df
  a b d e
1 1 1 x x
2 1 2 y y
3 2 2 z z
4 3 3 x x
5 4 4 x x
6 1 1 y x
7 1 1 x z
8 3 3 x x

If I apply this function df2 <- getRepplicate(df, allCol = c("a", "b"), anyCol = c("d", "e")), my expected result will be:

> df2
  a b d e isReplicate
1 1 1 x x TRUE
2 1 2 y y FALSE
3 2 2 z z FALSE
4 3 3 x x TRUE
5 4 4 x x FALSE
6 1 1 y x TRUE
7 1 1 x z TRUE
8 3 3 x x TRUE

Thanks for your help.

>Solution :

Is something like this?

In allCond we check if the number of unique values is one, so all values are the same
In anyCond we check if the number of unique values is equal to the number of values in the row
If both conditions are TRUE then it is replicated.

library(dplyr)

df <- data.frame(
  a = c(1, 1, 2, 3, 4, 1, 1, 3,1),
  b = c(1, 2, 2, 3, 4, 1, 1, 3,2), 
  d = c("x", "y", "z", "x", "x", "y", "x", "x","x"),
  e = c("x", "y", "z", "x", "x", "x", "z", "x","y")
)

getRepplicate <- function(df, allCol, anyCol) {
    df %>% 
    rowwise() %>% 
    mutate(
      allCond = n_distinct(c_across(all_of(allCol))) == 1 ,
      anyCond = n_distinct(c_across(all_of(anyCol))) < length(c_across(all_of(anyCol))) ) %>%
    ungroup() %>% 
    mutate(isReplicated = allCond & anyCond)
}
getRepplicate(df, allCol = c("a", "b"), anyCol = c("d", "e"))
#> # A tibble: 9 × 7
#>       a     b d     e     allCond anyCond isReplicated
#>   <dbl> <dbl> <chr> <chr> <lgl>   <lgl>   <lgl>       
#> 1     1     1 x     x     TRUE    TRUE    TRUE        
#> 2     1     2 y     y     FALSE   TRUE    FALSE
#> 3     2     2 z     z     TRUE    TRUE    TRUE        
#> 4     3     3 x     x     TRUE    TRUE    TRUE        
#> 5     4     4 x     x     TRUE    TRUE    TRUE        
#> 6     1     1 y     x     TRUE    FALSE   FALSE
#> 7     1     1 x     z     TRUE    FALSE   FALSE
#> 8     3     3 x     x     TRUE    TRUE    TRUE        
#> 9     1     2 x     y     FALSE   FALSE   FALSE

^{Created on 2023-02-10 with reprex v2.0.2}