CODEWITHSUNDEEP
rdata.table
Aoi Summer
Aoi Summer

Reputation: 35

Apply function on each cell in a data.frame

I have a data.frame like this:

G5_01
   X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
1:                                  0/0                                  0/0
2:                                  0/0                                  1/1
3:                                  0/1                                  0/0

I want to calculate variations in each cell and convert it to:

  X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
1:                                  0                                  0
2:                                  0                                  2
3:                                  1                                  0

data.table seemed to be able to deal with this and my script as shown below:

library(data.table)

G5_02<-setDT(G5_01)[,lapply(.SD,function(x) sum(as.numeric(strsplit(x,"/")[[1]][1]),
                                                as.numeric(strsplit(x,"/")[[1]][2])))]

But it only gives me the result of the first row

   X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
1:                                    0                                    0

Any suggestions on how should I fix it?

Upvotes: 0

Views: 106

Answers (5)

sm925
sm925

Reputation: 2678

library(data.table)
setDT(G5_01)[, X8803713069_R01C02_8803713069_R01C02 := as.numeric(substr(X8803713069_R01C02_8803713069_R01C02, 1, 1)) + as.numeric(substr(X8803713069_R01C02_8803713069_R01C02, 3, 3))][, X8803713069_R02C02_8803713069_R02C02 := as.numeric(substr(X8803713069_R02C02_8803713069_R02C02, 1, 1)) + as.numeric(substr(X8803713069_R02C02_8803713069_R02C02, 3, 3))]

G5_01
         X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
    1:                                    0                                    0
    2:                                    0                                    2
    3:                                    1                                    0

Data

G5_01 <- read.table(text = 'X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
                                  0/0                                  0/0
                                  0/0                                  1/1
                                 0/1                                  0/0', header = T)

Upvotes: 1

IceCreamToucan
IceCreamToucan

Reputation: 28705

Using data.table::tstrsplit to get a transposed string-split (like e.g. purrr::transpose(strsplit(x, '/'))) we can then convert to numeric and add them together

library(dplyr)

df %>% 
  mutate_all(~ 
    data.table::tstrsplit(., '/') %>% 
      map(as.numeric) %>% 
      do.call(what = '+'))

#   X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
# 1                                    0                                    0
# 2                                    0                                    2
# 3                                    1                                    0

Upvotes: 1

ThomasIsCoding
ThomasIsCoding

Reputation: 102625

Maybe you can try the code below, where nchar() and gsub() are used

Here are two solutions with base R:

  • Solution 1: (might be faster than sapply(), apply() or lapply() approaches)
G5_02 <- data.frame(nchar(gsub("[^1]","",as.matrix(G5_01))))
  • Solution 2:
G5_02 <- data.frame(sapply(G5_01, function(x) nchar(gsub("[^1]","",x))))

such that

> G5_01
  X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
1                                    0                                    0
2                                    0                                    2
3                                    1                                    0

DATA

G5_01 <- structure(list(X8803713069_R01C02_8803713069_R01C02 = c("0/0", 
"0/0", "0/1"), X8803713069_R02C02_8803713069_R02C02 = c("0/0", 
"1/1", "0/0")), class = "data.frame", row.names = c(NA, -3L))

Upvotes: 1

Sotos
Sotos

Reputation: 51592

If you only have to handle 1s and 0s, then a possible solution can be to count the 1s, i.e.

library(data.table)
setDT(df)[, lapply(.SD, function(i)stringr::str_count(i, '1'))][]

#   X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
#1:                                    0                                    0
#2:                                    0                                    2
#3:                                    1                                    0

Upvotes: 2

akrun
akrun

Reputation: 887831

In base R, we can use rowSums after doing the splitting with read.table

df[] <- lapply(df, function(x) rowSums(read.table(text = x,
        sep="/", header = FALSE)))
df
#  X8803713069_R01C02_8803713069_R01C02 X8803713069_R02C02_8803713069_R02C02
#1                                    0                                    0
#2                                    0                                    2
#3                                    1                                    0

data

df <- structure(list(X8803713069_R01C02_8803713069_R01C02 = c("0/0", 
"0/0", "0/1"), X8803713069_R02C02_8803713069_R02C02 = c("0/0", 
"1/1", "0/0")), class = "data.frame", row.names = c(NA, -3L))

Upvotes: 1

Related Questions