在尝试将值与 R 中的另一个对象匹配的函数上使用 lapply

Question

我希望有人能帮助我。我在 output

中有一份 ID 和血液结果列表

dput(输出)

list(BNP = structure(list(record_id = structure(c("113-1", "113-10", 
"113-11", "113-12", "113-13", "113-14"), label = c(record_id = "Record ID"), class = c("labelled", 
"character")), BNP = c(67.8, 1873.3, 784.5, 82.3, 156.5, 116.4
)), row.names = c(NA, -6L), class = c("tbl_df", "tbl", "data.frame"
)), proBNP = structure(list(record_id = structure(c("103-13", 
"103-16", "103-17", "103-20", "104-1", "106-12", "106-13", "106-14", 
"106-2", "106-3", "106-4", "106-6", "106-7", "106-8", "112-1"
), label = c(record_id = "Record ID"), class = c("labelled", 
"character")), proBNP = c(66, 1865, 6143, 275, 1240, 992, 1116, 
8469, 165, 2486, 85, 100, 9231, 8561, 815)), row.names = c(NA, 
-15L), class = c("tbl_df", "tbl", "data.frame")), troponin = structure(list(
    record_id = structure(c("101-2", "103-13", "103-14", "103-16", 
    "103-17", "104-1", "104-2", "106-12", "106-13", "106-14", 
    "106-5", "106-6", "106-7", "106-8", "112-1", "113-1", "113-10", 
    "113-11", "113-12", "113-13", "113-14"), label = c(record_id = "Record ID"), class = c("labelled", 
    "character")), troponin = c(29900, 5, 3, 5, 422, 11300, 10, 
    5.1, 1.5, 159.7, 15.9, 1863, 92.8, 706.5, 643, 50, 110, 60, 
    30, 130, 10)), row.names = c(NA, -21L), class = c("tbl_df", 
"tbl", "data.frame")), CRP = structure(list(record_id = structure(c("101-2", 
"103-13", "103-14", "103-15", "103-16", "103-17", "103-19", "103-20", 
"104-1", "104-2", "106-1", "106-11", "106-12", "106-13", "106-14", 
"106-2", "106-3", "106-4", "106-5", "106-6", "106-7", "106-8", 
"112-1", "113-1", "113-10", "113-11", "113-12", "113-13", "113-14"
), label = c(record_id = "Record ID"), class = c("labelled", 
"character")), CRP = c(54.8, 78, 229, 166, 77, 345, 25, 124, 
225.4, 156.2, 141.11, 110, 96.87, 126, 238.97, 6.19, 135.7, 135, 
138.7, 84.7, 242, 299, 41, 114.7, 156.2, 112.3, 394.3, 179.5, 
93)), row.names = c(NA, -29L), class = c("tbl_df", "tbl", "data.frame"
)), ferritin = structure(list(record_id = structure(c("101-2", 
"103-13", "103-14", "103-15", "103-16", "103-17", "103-20", "104-1", 
"106-11", "106-12", "106-13", "106-14", "106-2", "106-3", "106-4", 
"106-5", "106-6", "106-8", "112-1", "113-10", "113-11", "113-12", 
"113-13", "113-14"), label = c(record_id = "Record ID"), class = c("labelled", 
"character")), ferritin = c(253.97, 314, 438, 199, 390, 1342, 
128, 462.6, 125.8, 428, 237, 302.23, 1651, 133.6, 167, 1746, 
343, 1145.96, 697.76, 690.2, 395.4, 1492.2, 275.4, 254.2)), row.names = c(NA, 
-24L), class = c("tbl_df", "tbl", "data.frame")))

我在一个名为 identity

的单独 df 中也有一个 ID 列表

dput(身份)

structure(list(identity = c("101-2", "103-13", "103-14", "103-15", 
"103-16", "103-17", "103-19", "103-20", "104-1", "104-2", "106-1", 
"106-11", "106-12", "106-13", "106-14", "106-2", "106-3", "106-4", 
"106-5", "106-6", "106-7", "106-8", "112-1", "113-1", "113-10", 
"113-11", "113-12", "113-13", "113-14")), row.names = c(NA, -29L
), class = "data.frame")

我有一个血液测试向量，我试图通过它应用我的函数 key:

c("BNP", "proBNP", "troponin", "CRP", "ferritin")

key中的每个变量对应于其中一项血液测试。我正在尝试遍历 output 列表中的每个血液测试，然后将这些结果和相应的 ID 与 identity 数据框进行匹配，为每个血液测试创建一个新列。我认为问题可能是循环列表然后尝试输出到对象？我可能过于复杂了！

我失败的尝试：

#match output values with IDs
match_IDs <- function(x, y) {

#output[[ c(deparse(substitute(x)), "record_id") ]] gives record ID for blood test
#output[[ c(deparse(substitute(x)), deparse(substitute(x))) ]] gives blood value for blood test

y$new <- output[[ c(deparse(substitute(x)), deparse(substitute(x))) ]][match(y$`df2$record_id`, output[[ c(deparse(substitute(x)), "record_id") ]] )]
return(y$new)
}

lapply(key, identity, function(x) do.call("match_IDs", list(as.name(x)))) -> output2

在 match_IDs 函数之外，匹配代码有效，但我想自动执行该过程而不是硬编码。这就是我的目标：

structure(list(identity = c("101-2", "103-13", "103-14", "103-15", 
"103-16", "103-17", "103-19", "103-20", "104-1", "104-2", "106-1", 
"106-11", "106-12", "106-13", "106-14", "106-2", "106-3", "106-4", 
"106-5", "106-6", "106-7", "106-8", "112-1", "113-1", "113-10", 
"113-11", "113-12", "113-13", "113-14"), baseline_CRP = c(54.8, 
78, 229, 166, 77, 345, 25, 124, 225.4, 156.2, 141.11, 110, 96.87, 
126, 238.97, 6.19, 135.7, 135, 138.7, 84.7, 242, 299, 41, 114.7, 
156.2, 112.3, 394.3, 179.5, 93), baseline_bnp = c(NA, NA, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
NA, NA, NA, NA, 67.8, 1873.3, 784.5, 82.3, 156.5, 116.4), baseline_ferritin = c(253.97, 
314, 438, 199, 390, 1342, NA, 128, 462.6, NA, NA, 125.8, 428, 
237, 302.23, 1651, 133.6, 167, 1746, 343, NA, 1145.96, 697.76, 
NA, 690.2, 395.4, 1492.2, 275.4, 254.2), baseline_trop = c(29900, 
5, 3, NA, 5, 422, NA, NA, 11300, 10, NA, NA, 5.1, 1.5, 159.7, 
NA, NA, NA, 15.9, 1863, 92.8, 706.5, 643, 50, 110, 60, 30, 130, 
10), baseline_proBNP = c(NA, 66, NA, NA, 1865, 6143, NA, 275, 
1240, NA, NA, NA, 992, 1116, 8469, 165, 2486, 85, NA, 100, 9231, 
8561, 815, NA, NA, NA, NA, NA, NA)), row.names = c(NA, -29L), class = c("tbl_df", 
"tbl", "data.frame"))

如果有人有什么建议，我很想学习！

Answer 1

如果您修复 output，使其具有与 identity 相同的列名和类型，我们就可以进行一系列连接。

library(dplyr)
library(purrr)

# first, make the list fit the target
output_fixed <- map(
  output, 
  ~ .x %>% mutate(identity = as.character(record_id)) %>% select(-record_id)
)

# then, join repeatedly until at end of the list
output2 <- reduce(output_fixed, left_join, .init = identity, by = 'identity')

head(output2)

  identity BNP proBNP troponin   CRP ferritin
1    101-2  NA     NA    29900  54.8   253.97
2   103-13  NA     66        5  78.0   314.00
3   103-14  NA     NA        3 229.0   438.00
4   103-15  NA     NA       NA 166.0   199.00
5   103-16  NA   1865        5  77.0   390.00
6   103-17  NA   6143      422 345.0  1342.00

在尝试将值与 R 中的另一个对象匹配的函数上使用 lapply

Using lapply on a function that attempts to match values to another object in R

r

dataframe

function

dplyr

lapply