convert a list to data frame without messing with data-CodePudding

I have a list data like this

library(Matrix)
   
myd<- new("dgTMatrix", i = c(3L, 4L, 4L), j = 0:2, Dim = c(5L, 3L), 
        Dimnames = list(c("Xkr4", "Rp1", "Sox17", "Mrpl15", "Lypla1"
        ), c("AAAGTAGAGATGCCAG-1", "AACCGCGTCCAACCAA-1", "AACTCCCGTCGGGTCT-1"
        )), x = c(1, 2, 1), factors = list())

I am more comfortable to work with data frame than list, so I am trying to convert it to a data frame but I am not sure if I am destroying the structure or not

I simply do

mydf <- as.data.frame(myd)

then it becomes like this

mydf<- structure(list(`AAAGTAGAGATGCCAG-1` = c(0, 0, 0, 1, 0), `AACCGCGTCCAACCAA-1` = c(0, 
0, 0, 0, 2), `AACTCCCGTCGGGTCT-1` = c(0, 0, 0, 0, 1)), class = "data.frame", row.names = c("Xkr4", 
"Rp1", "Sox17", "Mrpl15", "Lypla1"))

Do you think there is a safer way to convert such a data?

I tried to understand the data, so if I unlist it, it will look like below, but still I am not sure of converting

unlist(myd)
5 x 3 sparse Matrix of class "dgTMatrix"
       AAAGTAGAGATGCCAG-1 AACCGCGTCCAACCAA-1 AACTCCCGTCGGGTCT-1
Xkr4                    .                  .                  .
Rp1                     .                  .                  .
Sox17                   .                  .                  .
Mrpl15                  1                  .                  .
Lypla1                  .                  2                  1

CodePudding user response：

There are a couple issues here. First off, calling as.data.frame(myd) where myd is an object with class dgTMatrix throws an error as shown below:

library(Matrix)

myd<- new("dgTMatrix", i = c(3L, 4L, 4L), j = 0:2, Dim = c(5L, 3L), 
          Dimnames = list(c("Xkr4", "Rp1", "Sox17", "Mrpl15", "Lypla1"
          ), c("AAAGTAGAGATGCCAG-1", "AACCGCGTCCAACCAA-1", "AACTCCCGTCGGGTCT-1"
          )), x = c(1, 2, 1), factors = list())

as.data.frame(myd)
#> Error in as.data.frame.default(myd): cannot coerce class 'structure("dgTMatrix", package = "Matrix")' to a data.frame

The reason for this is that there is no as.data.frame method defined for dgTMatrix. However, there is an as.matrix method defined for dgTMatrix as well as an as.data.frame method defined for class matrix. So, if your best course of action really is to be transitioning between dgTMatrix and data.frame (which I feel it may not be), you probably want to do something like:

library(Matrix)
   
myd<- new("dgTMatrix", i = c(3L, 4L, 4L), j = 0:2, Dim = c(5L, 3L), 
        Dimnames = list(c("Xkr4", "Rp1", "Sox17", "Mrpl15", "Lypla1"
        ), c("AAAGTAGAGATGCCAG-1", "AACCGCGTCCAACCAA-1", "AACTCCCGTCGGGTCT-1"
        )), x = c(1, 2, 1), factors = list())

mydf <- as.data.frame(as.matrix(myd))
mydf
#>        AAAGTAGAGATGCCAG-1 AACCGCGTCCAACCAA-1 AACTCCCGTCGGGTCT-1
#> Xkr4                    0                  0                  0
#> Rp1                     0                  0                  0
#> Sox17                   0                  0                  0
#> Mrpl15                  1                  0                  0
#> Lypla1                  0                  2                  1