melt to two variable columns

Question

I have the following variables in a data frame:

[1] \"Type\"   \"I.alt\"  \"idx06\"  \"idx07\"  \"idx08\" \"farve1\" \"farve2\"

If I do:

Answer 1

This doesn't answer your question about cast, but you could certainly subset and do two melts, followed by a merge:

dm1 <- melt(d[,c("Type","I.alt","idx06","idx07","idx08")], id=c("Type","I.alt"))
dm2 <- melt(d[,c("Type","I.alt","farve1","farve2")], id=c("Type","I.alt"))
colnames(dm2) <- c("Type", "I.alt", "variable2", "value2")
dm <- merge(dm1, dm2)

Or, equivalently, do one melt (like you're currently doing) then subset the melted dataframe twice (idx <- variable %in% c("idx06","idx07","idx08") as one and !idx as the other) and merge that output.

Either way you get what you want:

> head(dm)
              Type I.alt variable    value variable2 value2
1 Alvorligere vold  1154    idx08 1.108696    farve1    red
2 Alvorligere vold  1154    idx08 1.108696    farve2    red
3 Alvorligere vold  1154    idx07 1.027174    farve1    red
4 Alvorligere vold  1154    idx07 1.027174    farve2    red
5 Alvorligere vold  1154    idx06 1.000000    farve1    red
6 Alvorligere vold  1154    idx06 1.000000    farve2    red