R语言如何移除列表内数据框中D列含NA的同Study_ID全部行

现有存储数据框的R列表对象my.list，列表内每个数据框包含Study_ID、B、C、D四个字段，每个Study_ID对应2条记录。需要按如下规则清洗数据：

• 若任意数据框中，同一个Study_ID分组下存在任意一行的D列取值为NA，则将该Study_ID对应的全部记录从列表所有数据框中移除。

可复现测试数据构造代码如下：

my.list <- structure(list(S1 = structure(list(Study_ID = c(100, 100, 200, 
200, 300,300,400,400), B = c(NA, 1.5, 1.8, 2.1, 3.2, 1.4, NA, 9.3), C = c("C1", "PTA", "C1", "PTA", "C1", "PTA","C1", "PTA"), D = c(0.9124, NA, 0.5571429, 0.7849462, 0.32719, NA, 0.82482, 0.284702
)), .Names = c("Study_ID", "B", "C", "D"), class = "data.frame", row.names = c("1", 
"2", "3", "4", "5", "6", "7", "8")), S2 = structure(list(Study_ID = c(100, 100, 200, 
200, 300,300,400,400), B = c(NA, 0.7, NA, 0.45, 
0.91, 0.78, 0.65, NA), C = c("C1", "PTA", "C1", "PTA", "C1", "PTA", "C1", "PTA"), D = c(0.9124, NA, 0.5571429, 0.7849462, 0.32719,0.6492, 0.82482, NA
)), .Names = c("Study_ID", "B", "C", 
"D"), class = "data.frame", row.names = c("1", "2", "3", "4", 
"5", "6", "7", "8"))), .Names = c("S1", "S2"))

核心逻辑分两步：

1. 遍历列表所有数据框，汇总所有存在D列NA值对应的Study_ID，作为需要剔除的ID集合
2. 再次遍历列表所有数据框，保留Study_ID不在剔除集合中的行即可

基础R实现（无需加载第三方包） #

# 汇总需要剔除的Study_ID
exclude_ids <- unique(unlist(lapply(my.list, function(df) df$Study_ID[is.na(df$D)])))

# 过滤所有数据框
result <- lapply(my.list, function(df) {
  res_df <- df[!df$Study_ID %in% exclude_ids, ]
  # 重置行名，和预期输出格式匹配
  rownames(res_df) <- seq_len(nrow(res_df))
  res_df
})

tidyverse实现 #

如果习惯用tidyverse系列包操作，可以用如下代码：

library(dplyr)
library(purrr)

# 汇总需要剔除的Study_ID
exclude_ids <- my.list %>%
  map_dfr(~ .x %>% filter(is.na(D)) %>% select(Study_ID)) %>%
  distinct() %>%
  pull(Study_ID)

# 过滤所有数据框
result <- my.list %>%
  map(~ .x %>% 
        filter(!Study_ID %in% exclude_ids) %>% 
        `rownames<-`(seq_len(nrow(.))))

运行完成后result即为目标结果，打印输出和预期结果完全一致：

print(result)

输出：

$S1
  Study_ID   B   C         D
1      200 1.8  C1 0.5571429
2      200 2.1 PTA 0.7849462
3      400  NA  C1 0.8248200
4      400 9.3 PTA 0.2847020

$S2
  Study_ID    B   C         D
1      200   NA  C1 0.5571429
2      200 0.45 PTA 0.7849462
3      300 0.91  C1 0.3271900
4      300 0.78 PTA 0.6492000

内容的提问来源于stack exchange，提问作者sabc04