Combine dplyr mutate function with a search through the whole table - r

Question

I'm quite new to R and especially to the tidy verse. I'm trying to write a script with which we can rewrite a list of taxons. We already have one using quite a lot for and if loops and I want to try to simplify it with the tidyverse, but I'm kind of stuck how to do that.
what I have is a table that looks something like that (really simplified)
taxon_file<- tibble(name = c( "cockroach","cockroach2", "grasshopper", "spider", "lobster", "insect", "crustacea", "arachnid"),
Id = c(445,448,446,778,543,200,400,300),
parent_ID = c(200,200,200,300,400,200,400,300),
rank = c("genus","genus","genus","genus","genus","order","order","order")
)
+-------------+-----+-----------+----------+
| name | Id | parent_ID | rank |
+=============+=====+===========+==========+
| cockroach | 445 | 200 | genus |
| cockroach2 | 448 | 200 | genus |
| grasshopper | 446 | 200 | genus |
| spider | 778 | 300 | genus |
| lobster | 543 | 400 | genus |
| insect | 200 | 200 | order |
| crustacea | 400 | 400 | order |
| arachnid | 300 | 300 | order |
+-------------+-----+-----+------------+----------+
Now I want to rearrange it so that I get a new column where I can add the order that matches the parent_ID (so when parent_ID == ID then write name in column order). The end result should look kinda like this
+-------------+------------+------+-----------+
| name | order | Id | parent_ID |
+=============+============+======+===========+
| cockroach | insect | 445 | 200 |
| cockroach2 | insect | 448 | 200 |
| grasshopper | insect | 446 | 200 |
| spider | arachnid | 778 | 300 |
| lobster | crustacea | 543 | 400 |
+-------------+------------+------+-----------+
I tried to combine mutate with an ifelse statement but this just adds NA's to the whole order column.
tibble is named taxon_list
taxon_list %>%
mutate(order = ifelse(parent_ID == Id, Name, NA))
I know this will not work because it doesn't search the whole data-set for the correct row (that's what I did before with alle the for loops). Maybe someone can point me in the right direction?

Answer 1

One way is to filter to each rank type to 2 separate dfs, subset using select, and merge the 2.
df <- tibble(name = c( "cockroach","cockroach2", "grasshopper", "spider", "lobster", "insect", "crustacea", "arachnid"),
Id = c(445,448,446,778,543,200,400,300),
parent_ID = c(200,200,200,300,400,200,400,300),
rank = c("genus","genus","genus","genus","genus","order","order","order"))
library(tidyverse)
df_order <- df %>%
filter(rank == "order") %>%
select(order = name, parent_ID)
df_genus <- df %>%
filter(rank == "genus") %>%
select(name, Id, parent_ID) %>%
merge(df_order, by = "parent_ID")
Result:
parent_ID name Id order
1 200 cockroach 445 insect
2 200 cockroach2 448 insect
3 200 grasshopper 446 insect
4 300 spider 778 arachnid
5 400 lobster 543 crustacea