Revert split effect in a list

Question

I have a list and I would like to convert it into a data.frame with two columns. The problem is that the length of list elements is not equal, here is a example of how my data looks:

my.list

$A1CF
 [1] "A1CF"    "APOBEC1" "CUGBP2"  "KHSRP"   "SYNCRIP" "TNPO2"  

$A2LD1
[1] "A2LD1"   "PRPSAP2" "RPL15"   "TANC1"  

$A2M
[1] "A2M"      "ADAM19"   "ADAMTS1"  "AMBP"     "ANXA6"    "APOE"

This list comes from a previous data.frame:

my.list <- split(df$V2, df$V1) 

df
      V1      V2
1   A1BG    A1BG
2   A1BG  CRISP3
3   A1CF    A1CF
4   A1CF APOBEC1
5   A1CF  CUGBP2
6   A1CF   KHSRP
7   A1CF SYNCRIP
8   A1CF   TNPO2
9  A2LD1   A2LD1
10 A2LD1 PRPSAP2
11 A2LD1   RPL15
12 A2LD1   TANC1
13   A2M     A2M
14   A2M  ADAM19
15   A2M ADAMTS1
16   A2M    AMBP
17   A2M   ANXA6
18   A2M    APOE

where the elements corresponding to AB1G were removed. I would like to revert the split effect to obtain the same structure:

new.df
A1CF    A1CF
A1CF APOBEC1
A1CF  CUGBP2
A1CF   KHSRP
A1CF SYNCRIP
A1CF   TNPO2
A2LD1   A2LD1
A2LD1 PRPSAP2
A2LD1   RPL15
A2LD1   TANC1
A2M     A2M
A2M  ADAM19
A2M ADAMTS1
A2M    AMBP
A2M   ANXA6
A2M    APOE

I have tried with: df.new <- do.call(rbind, my.list), but obviously it didn't work.

Many thanks

Answer 1

Using these dummy data,

ll <- list(a = letters[3:6],
           b = LETTERS[1:10],
           c = letters[1:2])

stack(ll)

or

reshape2::melt(ll, id=1)

or

plyr::ldply(ll, cbind)

should give you roughly the right format

Answer 2

If you have a list of vectors, @baptiste provides more concise and general answers for list.

You can also apply unsplit, the inverse operation of split.

options(stringsAsFactors = FALSE)
df <- data.frame(V1 = c('A1BG', 'A1BG', 'A1CF', 'A1CF', 'A1CF', 'A1CF', 'A1CF', 'A1CF', 
                        'A2LD1', 'A2LD1', 'A2LD1', 'A2LD1', 'A2M', 'A2M', 'A2M', 'A2M', 
                        'A2M', 'A2M'),
                 V2 = c('A1BG', 'CRISP3', 'A1CF', 'APOBEC1', 'CUGBP2', 'KHSRP', 'SYNCRIP',
                        'TNPO2', 'A2LD1', 'PRPSAP2', 'RPL15', 'TANC1', 'A2M', 'ADAM19', 
                        'ADAMTS1', 'AMBP', 'ANXA6', 'APOE'))
my.list <- split(df$V2, df$V1)
newdat <- data.frame(V1=df$V1, V2=unsplit(my.list, df$V1))

If you have a list of data.frame from split, unsplit should be appropriate.

my.list <- split(df, df$V1)
newdf <- unsplit(my.list[!names(my.list) %in% 'A1BG'], df$V1[!df$V1 %in% 'A1BG']) 

or

newdf <- do.call(rbind, my.list[!names(my.list) %in% 'A1BG'])