6

I am trying to transpose a data frame in R, but having very little luck.

The data frame contains an epigenetic data set, with 300,000+ CpG sites in the first column. The 74 additional columns are split between the control and experimental groups (cancer = 69, normal = 5).

I have converted the data frame into a matrix so I can transpose the data and convert it back to numerical values. However, every time I try to convert the data back to a data frame, it ends up as a list.

I'm attempting to do the following:

  1. Make the first column the headers.

  2. Make the name of the headers the values in the first column (except for X, which I would like to remove).

    data[1:10, 1:10]

        X Tumor.33 Normal.01 Tumor.01 Tumor.34 Tumor.35 Tumor.02 Tumor.03


cg00000029  0.29224   0.32605  0.58762  0.32397  0.23482  0.24012  0.22941

cg00000108  0.91243   0.89785  0.92337  0.90080  0.91220  0.92256  0.92709

cg00000109  0.77676   0.73910  0.81545  0.73603  0.76276  0.85808  0.85142

cg00000165  0.34261   0.30960  0.56392  0.32363  0.33980  0.61755  0.70855

cg00000236  0.84688   0.80654  0.84423  0.80935  0.85600  0.87766  0.83509

cg00000289  0.61535   0.60874  0.62496  0.66421  0.60556  0.66824  0.65243

cg00000292  0.72491   0.63333  0.55031  0.69690  0.73547  0.71826  0.62223

cg00000321  0.31650   0.29422  0.37737  0.28428  0.28417  0.70437  0.34829

cg00000363  0.26309   0.31460  0.29135  0.26339  0.20117  0.60604  0.60548

cg00000622  0.03325   0.02190  0.03293  0.04032  0.02815  0.03494  0.04126
statsguyz
  • 419
  • 2
  • 11
  • 35

2 Answers2

7

data <- as.data.frame(t(data))

This ensures that the numeric values are not converted to string values.

Miha
  • 2,559
  • 2
  • 19
  • 34
statsguyz
  • 419
  • 2
  • 11
  • 35
6

Use the t() function to transpose a matrix or a data frame. In the latter case, row names become variable (column) names.

df > data[1:10, 1:10]

t(df)

Source: https://www.r-statistics.com/tag/transpose/

Community
  • 1
  • 1
jemman
  • 163
  • 1
  • 2
  • 12
  • Ok, thanks. I've had issues with the Beta values turning in to string values. How can I turn all of the data back to numeric values? – statsguyz Nov 13 '16 at 22:21
  • You're welcome. I found this answer that maybe can help you: http://stackoverflow.com/questions/6778908/r-transposing-a-data-frame. –  Nov 13 '16 at 22:55