how to change row values based on information from another dataframe in R-CodePudding

I have the original df:

A <- c("A1", "A2", "A3", "A4")
B <- c(1,0,1,NA)
C <- c(0,1,0,NA)
D <- c(NA, 1, 0, NA)
              
df <- data.frame(A, B, C, D)

And my second df2:

A <- c("A2", "A3")
df2 <- data.frame(A)

I would like to modify df_modified to look like this

A    B   C   D
A1   1   0   NA
A2   NA  NA  NA
A3   NA  NA  NA 
A4   NA  NA  NA

My current code, which generated all rows filled by NA is:

df_modifed <- df %>% mutate(B = case_when(df$A == df2$A ~ NA),
              C = case_when(df$A == df2$A ~ NA),
               D = case_when(df$A == df2$A ~ NA))

And assistance in that problem would be greatly appreciated!!

Thanks in advanced!!

CodePudding user response：

In base R, this is easier i.e. specify the logical index as row and column index without the first column (-1) and assign those elements to NA

df[df$A %in% df2$A, -1] <- NA

-output

> df
   A  B  C  D
1 A1  1  0 NA
2 A2 NA NA NA
3 A3 NA NA NA
4 A4 NA NA NA

Or if we want to use tidyverse, use across

library(dplyr)
df %>%
   mutate(across(where(is.numeric), ~ case_when(!A %in% df2$A~ .)))

-output

   A  B  C  D
1 A1  1  0 NA
2 A2 NA NA NA
3 A3 NA NA NA
4 A4 NA NA NA

CodePudding user response：

Here is an alternative dplyr way:

bind_rows(df, df2) %>% 
  group_by(A) %>% 
  mutate(across(c(B,C,D), ~first(.)==last(.))*1) %>% 
  distinct()

  A         B     C     D
  <chr> <dbl> <dbl> <dbl>
1 A1        1     1    NA
2 A2       NA    NA    NA
3 A3       NA    NA    NA
4 A4       NA    NA    NA