CODEWITHSUNDEEP
rtidyverse
Nico Biagi
Nico Biagi

Reputation: 21

Add an incremental counter to a dataframe in R

I have a data-frame composed of 3 columns and the third column ('V3') contains the name of markers that were presented during the experiment. I'd like to add another column that tells me for each marker how many time that specific marker was presented before that instance. I have done it in R without using the tidyverse but it is quite time consuming, and I was wondering if you could help me do it using the tidyverse.

At the moment my script goes like this:

data$counter <- NA
x<-1
for (i in 1:nrow(data)){
 if ((str_detect(data$V3[i], 'NEGATIVE'))==TRUE){
  data$counter[i] <- x
  x <- x +1
 }
}

Upvotes: 0

Views: 1463

Answers (3)

Jonas
Jonas

Reputation: 1810

I think cumsum will do an excellent job here:

testdata <- data.frame(V3=sample(c("NEGATIVEsomething","others"),50,replace=TRUE), stringsAsFactors = FALSE)

negatives <- grepl("NEGATIVE",testdata$V3)
negatives <- as.numeric(negatives)
negatives <- cumsum(negatives)
negatives[negatives == 0] <- NA

testdata$counter <- negatives

Edit: Since you want to increment the counter after finding a "NEGATIVE" and place the old counter at the position, you should use

negatives <- cumsum(negatives)-1

and then remove the 0 and -1 counts at the beginning:

negatives[negatives  %in% c(0,-1)] <- NA

Upvotes: 1

Regis
Regis

Reputation: 148

Try

data$counter <- cumsum((str_detect(data$V3, 'NEGATIVE')))

Upvotes: 0

Ben Norris
Ben Norris

Reputation: 5747

You can do this with group_by() and then generating the counters with seq_along().

library(dplyr)

data %>%
  group_by(V3) %>%
  mutate(counter = seq_along(V3)) %>%
  ungroup()

Upvotes: 1

Related Questions