I have a data frame that looks like this :

If the values of cond1 and cond2 columns is 0 then 0 must be the columns a and b and then 1 for columns c and d.

Ideally I want a resulted data frame like this :

How I can do this in R using dplyr package ?

library(tidyverse)
cond1 = c(1,1,0,0,1,1)
cond2 = c(0,0,0,0,1,1)
a = c(98,45,34,21,1,34)
b = c(89,54,233,12,1,4)
c = c(78,34,56,4,2,123)
d = c(89,56,67,78,89,2)
data = tibble(cond1,cond2,a,b,c,d);data

CodePudding user response：

With dplyr, if there are separate values to be replaced for blocks of columns, call the across with those subset of columns and replace

library(dplyr)
data <- data %>%
   mutate(across(c(a,b), ~  .x * !(cond1==0 & cond2==0)), 
    across(c(c, d), ~ replace(.x,(cond1 == 0 & cond2 == 0), 1)))

-output

data
# A tibble: 6 × 6
  cond1 cond2     a     b     c     d
  <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1     1     0    98    89    78    89
2     1     0    45    54    34    56
3     0     0     0     0     1     1
4     0     0     0     0     1     1
5     1     1     1     1     2    89
6     1     1    34     4   123     2

Or use a single across and create a condition with case_when

data %>% 
  mutate(across(a:d, ~ case_when(cond1 == 0 & cond2 == 0 ~ c(0, 1)[
      1   any(c('c', 'd') %in% cur_column())], TRUE ~ .x)))

-output

# A tibble: 6 × 6
  cond1 cond2     a     b     c     d
  <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1     1     0    98    89    78    89
2     1     0    45    54    34    56
3     0     0     0     0     1     1
4     0     0     0     0     1     1
5     1     1     1     1     2    89
6     1     1    34     4   123     2