Select from data.frame based on another data.frame in R

Question

I have a very simple question I struggling to solve in R (find many answers in other coding systems).

I have a data.frame with an ID field with several IDs:

> data_new <- data.frame(ID_ornitho = c("1344", "2364", "1111","2254"))
> data_new
  ID_ornitho
1       1344
2       2364
3       1111
4       2254

I have another data.frame with ID's already used:

> data_old <- data.frame(ID_ornitho = c("2354", "2364", "2254","1354"))
> data_old
  ID_ornitho
1       2354
2       2364
3       2254
4       1354

What I would like to do is simple to delete from data_new the rows corresponding to ID's already used in data_old , achieving this:

> data_filtered
  ID_ornitho
1       1344
2       1111

So simple that I cannot find a simple way to do it!

Answer 1

You can use dplyr to filter for existing ids:

library(dplyr)
data_old <- data.frame(ID_ornitho = c("2354", "2364", "2254","1354"))
data_new <- data.frame(ID_ornitho = c("1344", "2364", "1111","2254"))
data_new %>% filter(!(ID_ornitho %in% data_old$ID_ornitho))

This gives

data_new %>% filter(!(ID_ornitho %in% data_old$ID_ornitho))
  ID_ornitho
1       1344
2       1111

Answer 2

Staying in base, you can use a logical vector to subset data_new like so:

data.frame(ID_ornitho=
               data_new[!data_new$ID_ornitho %in% data_old$ID_ornitho, ])

See ? match ? match for details and more examples.

Answer 3

Perfect usecase for anti_join from dplyr package:

library(dplyr)
anti_join(df1, df2, by="ID_ornitho")

  ID_ornitho
1       1344
3       1111

Answer 4

You can easily use library(dplyr)

anti_join(x,y by="ID_ornitho", copy= False)