apply function to certain columns of all dataframe in list and then assign value to columns

问题

Similar question is answered here

I have a list of data frames (i.e. 1000) that looks like this:

> lst[1000]
$rand1000
                        Binomial         S4         S5    S6
254            Euastacus suttoni  25.816165  4.1916527  86.3
327            Orconectes hobbsi  16.726409  5.5241308  25.0
282           Faxonella creaseri  17.026970  6.4109494  18.0
319           Orconectes durelli  14.093957  7.2342324  35.0
525           Procambarus raneyi  15.799001  6.3746574  55.1

I want to apply function to S4, S5 and S6 columns for all data frames in the list. So I wrote this function:

lapply(lst, function(x) {x$S4 <- sensitivity.rand(x[[2]], 25); x})

(here 'sensitivity.rand' is a function that takes the vector and threshold (in the above code 25 is the threshold) and assign into H or L)

that give the output as (which is good):

$rand1000
                        Binomial S4         S5    S6
254            Euastacus suttoni  H  4.1916527  86.3
327            Orconectes hobbsi  H  5.5241308  25.0
282           Faxonella creaseri  H  6.4109494  18.0
319           Orconectes durelli  H  7.2342324  35.0
525           Procambarus raneyi  H  6.3746574  55.1

But if I see that the original data frame is not changed (i.e. it is as it was before). How can I do this. I need to do this for all S4, S5 and S6 columns for all data frame in the list. So that the original data frames will change into like this:

> lst[1000]
$rand1000
                            Binomial S4  S5 S6
    254            Euastacus suttoni  H  H  H
    327            Orconectes hobbsi  H  L  H
    282           Faxonella creaseri  H  H  L
    319           Orconectes durelli  H  L  L
    525           Procambarus raneyi  H  H  H

回答1:

We need to assign the output back to the list or create a new object. Also, in the OP's code, the function is only applied to the 2nd column. We can either loop over the columns of interest and apply the function or do it separately

lst <- lapply(lst, function(x) {x[2:4] <- lapply(x[2:4], sensitivity.rand, threshold = 25)
                 x})

Or if we are using tidyverse, this can be done with mutate_each

lst <- lapply(lst, function(x) x %>%
                                 mutate_each(funs(sensitivity.rand(., 25)), 2:4))

来源：https://stackoverflow.com/questions/43059050/apply-function-to-certain-columns-of-all-dataframe-in-list-and-then-assign-value

标签

list

dataframe

lapply