Create lists with values of comparisons between groups stored within a table in R-CodePudding

I have a dataframe such as ;

Comparison Distance
A,B        2
A,C        30
A,D        32
A,E        34
B,C        31  
B,D        40
B,E        5
C,E        2
C,D        6
D,E        7

and I have two list such as ;

Group_1 <- c("A","B")
Group_2 <- c("C","D","E")

and I would like to generate three new lists where I put all the Comparison between the element in Group_1 and Group_2 in a List1_G1vsG2, one List2_G1vsG1 with comparison within element of Group_1 and on in List3_G2vsG2 for comparison within elements of Group_1.

I shoul then get the following lists:

List1_G1vsG2

c(30,32,34,31,40)

List2_G1vsG1

c(2)

List3_G2vsG2

c(5,2,6,7)

Here is the dataframe in dput() format if it can help

structure(list(Comparison = c("A,B", "A,C", "A,D", "A,E", "B,C", 
"B,D", "B,E", "C,E", "C,D", "D,E"), Distance = c(2, 30, 32, 34, 
31, 40, 5, 2, 6, 7)), class = "data.frame", row.names = c(NA, 
-10L))

CodePudding user response：

A possible solution:

library(tidyverse)

df %>% 
  separate(Comparison, into = c("C1", "C2")) %>% 
  mutate(G = case_when(C1 %in% Group_1 & C2 %in% Group_1 ~ 1, 
                       C1 %in% Group_2 & C2 %in% Group_2 ~ 2,
                       C1 %in% Group_1 & C2 %in% Group_2 ~ 3)) %>% 
  group_by(G) %>% group_split %>% 
  map(pull, Distance) %>% set_names(c("List2_G1vsG1", "List3", "List1_G1vsG2"))

#> $List2_G1vsG1
#> [1] 2
#> 
#> $List3
#> [1] 2 6 7
#> 
#> $List1_G1vsG2
#> [1] 30 32 34 31 40  5

CodePudding user response：

Maybe this is beyond the scope of your request. You can customize a distance matrix computation function. By this way, you can determine whether or not the distance between A and B is equal to that of B and A.

f <- function(x, y, dist.df){
  pair <- outer(x, y, paste, sep = ",")
  dist <- dist.df[[2]][match(pair, dist.df[[1]])]
  matrix(dist, length(x), dimnames = list(x, y))
}

f(Group_1, Group_2, df)
#    C  D  E
# A 30 32 34
# B 31 40  5

f(Group_1, Group_1, df)
#    A  B
# A NA  2
# B NA NA

f(Group_2, Group_2, df)
#    C  D  E
# C NA  6  2
# D NA NA  7
# E NA NA NA