This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Unlisting lists inside data frame and collapsing their values in R

Hello I have this data.frame and as you can see there are lists inside some cells.

myList1 <- list()
myList1[[1]] <- 0
myList1[[2]] <- list(3)
myList1[[3]] <- list(6)
myList1[[4]] <- list(7, 9)

myList <- list()
myList[[1]] <- list(1, 4, 6, 7)
myList[[2]] <- list(2, 7, 3)
myList[[3]] <- list(5, 5, 3, 9, 6)
myList[[4]] <- list(7, 9)

myDataFrame <- data.frame(row = c(1,2,3,4))

myDataFrame$col1 <- myList1
myDataFrame$col2 <- myList

the data frame looks like:

row   col1          col2
   1    0           list(1, 4, 6, 7)
   2    list(3)     list(2, 7, 3)
   3    list(6)     list(5, 5, 3, 9, 6)
   4    list(7, 9)  list(7, 9)

Is there any way to unlist the lists and collapse with colon symbol, their items in order to make the dataframe look like the following ?

 row   col1           col2
   1    0             1:4:6:7
   2    3             2:7:3
   3    6             5:5:3:9:6
   4    7:9           7:9

Thank you

r lists unlist dataframe
• 13,961 views • link
updated by opplatek opplatek • written by arronar arronar

2 answers

My first thought is that you should examine your choices that got you to this point and choose a better data representation.

If you must fix this problem as described, then something like this should work:

for(i in 1:length(myDataFrame[,1])){
    if(is.list(myDataFrame[i,3])){
        myDataFrame[i,3]=paste(unlist(myDataFrame[i,3]),collapse=":")
    }
}
• 0 views • link
written by Chris Miller Chris Miller

If you don't want to use for loop you can just use a combination of apply (to iterate over columns) and sapply (to iterate over lists). 

# Apply to all columns
apply(myDataFrame, 2, function(y) sapply(y, function(x) paste(unlist(x), collapse=":")))

# Apply to all columns but the first one
apply(myDataFrame[, -1], 2, function(y) sapply(y, function(x) paste(unlist(x), collapse=":")))
• 0 views • link
written by opplatek opplatek

Log in to answer this question.

More posts like this