CODEWITHSUNDEEP
rlistdataframedata-manipulation
Kim So Yon
Kim So Yon

Reputation: 85

R: Creating a new column with desired alphabets

I am a beginner in R, so I am sorry if my question is too basic, but I would really appreciate some help in this.

  mydata <-
structure(list(Col1 = c(17, 28, 80, 63, 20, 
10), Col2 = c(18, 27, 89, 62, 24, 
11), Col3 = c(25, 40, 80, 65, 23, 
11), Col4 = c(27, 29, 100, 72, 34, 
6)), class = "data.frame", 
row.names = c("row1", "row2", "row3", "row4", "row5", 
"row6"))

I would like to add a new column 'X'. For 'X', I would like to assign A for Row 1-2, B for row 3-4, C for row 5 and D for row 6.

The code I tried is..

mydata$X[mydata[c(1:2),]]<-A
mydata$X[mydata[c(3:4),]]<-B
mydata$X[mydata[c(5),]]<-C
mydata$X[mydata[c(6),]]<-D

I tried putting "" e.g. "A" when I am assigning letters, but couldn't get it to work.

I got error message:

invalid subscript type 'list'

So, I tried unlisting my data, but still did not work.

Can anybody help please?

Upvotes: 0

Views: 268

Answers (2)

aiatay7n
aiatay7n

Reputation: 182

r2evans has answered the original question completely This is a new and unclear classification question: "I wanted to classify them into four different groups (A, T, C, G) according to the start of my sequences. That too appears to be answered by r2evans: mydata$X[1:2] <- "A" Extendable to: mydata$X <- c(rep("A",2), rep("B",2),rep("C",1),rep("D",1)) Ronak's recent answer of course is more eloquent!

Upvotes: 2

Ronak Shah
Ronak Shah

Reputation: 388982

You can use case_when from dplyr. We use grepl to detect the pattern based on start of sequence and assign values accordingly. 

library(dplyr)

mydata %>%
  #If the value starts with "AAT" assign "A"
  mutate(X = case_when(grepl('^AAT', column) ~ 'A', 
  #If the value starts with "ABC" assign "B"
                       grepl('^ABC', column) ~ 'B', 
                       #More cases
                       #More cases
  #If none of them satisfy assign `NA`
                       TRUE ~NA_character_))

Instead of grepl you can also use startsWith or str_detect.

Upvotes: 2

Related Questions