R:如何根据分组变量并以前面的行为条件来编码新变量

Question

我有一个器官移植数据集。数据是按捐赠者组织的，我正在查看肺部 - 所以每个捐赠者都有两个肺部。数据整理如下:

library(tidyverse)

data <- tribble(
  ~donor_id, ~sequence, ~organ_placed,
  1, 5, "L",
  1, 10, "R",
  2, 13, "B",
  3, 4, "L",
  3,69,NA,
  3,70,NA,
  3,71,NA,
  3, 72, NA,
)

• donor_id = 捐赠者 ID

• 序列 = 捐赠者报价的编号。 IE。它传给第 1 个人，然后是 2，然后是 3...等等。

• organ_placed = 如果器官匹配，则列出放置的器官。例如。对于供体 1，左肺放置在第 5 个供体上，右肺放置在第 10 个供体上。

我正在尝试弄清楚如何为捐赠者 3 编写案例:左肺被放置在第 3 个报价上，但右肺从未被放置 - 匹配报价持续到 72 个报价然后停下来。

我希望新数据如下所示:

desired_data <- tribble(
  ~donor_id, ~sequence, ~organ_placed,~outcome,
  1, 5, "L","Left Single",
  1, 10, "R","Right Single",
  2, 13, "B","Bilateral",
  3, 4, "L","Left Single",
  3,69,NA,NA,
  3,70,NA,NA,
  3,71,NA,NA,
  3, 72, NA,"Right Discarded"
)

我认为这类似于group_by(donor_id)，然后sequence的slice_max有NA organ_placed，但我需要它来编码它是被丢弃的右肺。

data <- data %>%
  mutate(outcome = case_when(
    organ_placed=="L" ~ "Left Single",
    organ_placed=="R" ~ "Right Single",
    organ_placed=="B" ~ "Bilateral",
    **what would go here? to group by donor id and find the maxslice has an NA, 
      and L (or R) already occurred in that donor?**
  ))

感谢您的帮助!

Answer 1

我们可以这样做: 基本上，在这种方法中，我们使用 case_when 简单地遍历所有条件:

library(dplyr)
data %>% 
  group_by(donor_id) %>% 
  mutate(outcome = case_when(organ_placed == "L" ~ "Left Single",
                            organ_placed == "R" ~ "Right Single",
                            organ_placed == "B" ~ "Bilateral",
                            (is.na(organ_placed) & 
                              row_number() == max(row_number())) & 
                              first(organ_placed) == "L" ~ "Right Discarded",
                            (is.na(organ_placed) & 
                               row_number() == max(row_number())) & 
                              first(organ_placed) == "R" ~ "Left Discarded",
                            TRUE ~ NA_character_))

  donor_id sequence organ_placed outcome        
     <dbl>    <dbl> <chr>        <chr>          
1        1        5 L            Left Single    
2        1       10 R            Right Single   
3        2       13 B            Bilateral      
4        3        4 L            Left Single    
5        3       69 NA           NA             
6        3       70 NA           NA             
7        3       71 NA           NA             
8        3       72 NA           Right Discarded