Remove variables with factor level 1

Question

I am using the program gs in the bnlearn package for my data frame EMGbin. The dataframe EMGbin contains all factors, ranging from A to Z. EMGbin has 600000 columns and 130 rows. Here is a sample of EMGbin:

   V101  V102  V103  V104  V105  V106
 1    L     M     D     S     O     O
 2    L     M     C     P     A     O
 3    J     M     C     O     O     O
 4    L     N     D     R     A     O
 5    K     M     D     O     A     O
 6    K     M     C     P     O     O
 7    K     N     D     Q     O     O
 8    L     N     D     R     O     O
 9    L     M     D     O     O     O
10    K     M     D     S     A     O

When I run the program gs(EMGbin), I get the error:

Error in check.data(x) : all factors must have at least two levels.

When I run sapply(EMGbin, nlevels), I see the levels of factors each of the 600,000 variables has, and I see some of them are listed as 1 level. Would removing the variables with 1 factor level help? So far, the only way I know how to do this is x[, sapply(x, fun) != 1], but I don't know what to substitute in for fun.

Answer 1

Use this:

x[, sapply(x, nlevels) > 1]

Answer 2

You can check the number of levels in a factor with the nlevels function.