在 R 的 data.frames 列表中使用一个 data.frame 的指定列

Question

I want to subtract the "response" column of a certain data.frame from a list of data.frames in R. Below is a subset of the data, and I want to use the "response" of B1.xls as a control, from which all the other "response" columns (A1.xls and A2.xls) are subtracted. Any suggestions to achieve it in a fast way? Thank you.

$A1.xls
    time   response
1 1320.0 0.00978786
2 1320.2 0.01236774
3 1320.4 0.01582583
4 1320.6 0.01947132
5 1320.8 0.02356580
6 1321.0 0.02786478

$A2.xls
    time   response
1 1320.0 0.00792504
2 1320.2 0.01079251
3 1320.4 0.01420215
4 1320.6 0.01848047
5 1320.8 0.02318325
6 1321.0 0.02836512

$B1.xls
    time   response
1 1320.0 0.00380097
2 1320.2 0.00515638
3 1320.4 0.00658969
4 1320.6 0.00803828
5 1320.8 0.00980306
6 1321.0 0.01187936

Answer 1

An option is lapply . Assuming that we want to only update the list elements with names that starts with "A", use lapply to loop over the subset of the list based on the grep on the names , then use transform to update the response by subtracting from the 'response' column from 'B1.xls' and assign it back to the subset list

nm1 <- grep("^A\\d+\\.xls", names(lst1))
#or
nm1 <- setdiff(names(lst1), "B1.xls")

lst1[nm1] <- lapply(lst1[nm1], transform,
          response = response - lst1$B1.xls$response)

NOTE: Here, we assume that all the list elements have the same number of rows

---

or we can use tidyverse approach with map

library(purrr)
library(dplyr)
library(stringr)
map_if(lst1, .p = str_detect(names(lst1), "^A\\d+\\.xls$"), ~
       .x %>%
             mutate(response = response - lst1$B1.xls$response))

data

lst1 <- list(A1.xls = structure(list(time = c(1320, 1320.2, 1320.4, 1320.6, 
1320.8, 1321), response = c(0.00978786, 0.01236774, 0.01582583, 
0.01947132, 0.0235658, 0.02786478)), class = "data.frame", row.names = c("1", 
"2", "3", "4", "5", "6")), A2.xls = structure(list(time = c(1320, 
1320.2, 1320.4, 1320.6, 1320.8, 1321), response = c(0.00792504, 
0.01079251, 0.01420215, 0.01848047, 0.02318325, 0.02836512)), class = "data.frame", row.names = c("1", 
"2", "3", "4", "5", "6")), B1.xls = structure(list(time = c(1320, 
1320.2, 1320.4, 1320.6, 1320.8, 1321), response = c(0.00380097, 
0.00515638, 0.00658969, 0.00803828, 0.00980306, 0.01187936)), 
class = "data.frame", row.names = c("1", 
"2", "3", "4", "5", "6")))

Answer 2

You can use purrr::update_list :

library(purrr)

lst1 %>%
  update_list(A1.xls = ~ A1.xls - B1.xls,
              A2.xls = ~ A2.xls - B1.xls)

Answer 3

A base R option using simple for loop

for (k in seq_along(lst)) {
  if (names(lst)[k] != "B1.xls") {
    lst[[k]]$response <- lst[[k]]$response - lst[["B1.xls"]]$response
  }
}

Answer 4

Using Map() :

control_df_name <- "B1.xls"

c(Map(function(x){x$response <- x$response - lst1[[control_df_name]]$response; x}, 
    lst1[names(lst1) != control_df_name]), lst1[control_df_name])[names(lst1)]