How to assign color scale to a variable in a 3D scatter plot?

Question

I'm a beginner in R and needs a bit of help for my scripting. I managed to generate scale color gradient using library(ggplot2) on my 2D plots as follows;

z <- c(data$conf)
d <- qplot(x, y, xlab="Dimension 1", ylab="Dimension 2", colour=z)
d
d + scale_colour_gradient(limits=c(0, 1), data=data$conf, low="blue", high="red"))

I'm now trying to reproduce this gradient on a 3D plot, I used scatterplot3d or plot3d. I believe the colorRampPalette create a color gradient based on 327 rows (1…327) while I'm interested in a gradient that is function of values in data$conf. I need a connection, but where?

attach(data)
t1 <- c(data$conf)
jet.colors <- colorRampPalette(c("blue", "red"))
e <- plot3d(x, y, z, col=jet.colors(327))

If you can help me that will be great – Or if you know any 3D plot/scale gradient package that can do a better job, cool too.

Answer 1

You are on the right track with colorRampPalette(), but really need something more like colorRamp(), which 'returns a function that maps values between 0 and 1'.

Even better would be a function -- call it myColorRamp() -- which is like colorRamp() but instead: (a) maps values between min(values) and max(values); and (b) returns the colors as 7-character sRGB strings (e.g. "#F60008"), a format that plot3d() understands.

library(rgl)

myColorRamp <- function(colors, values) {
    v <- (values - min(values))/diff(range(values))
    x <- colorRamp(colors)(v)
    rgb(x[,1], x[,2], x[,3], maxColorValue = 255)
}

x <- sin((1:100)/10)
y <- cos((1:100)/10)
z <- seq(-20, 20, length.out=100)

cols <- myColorRamp(c("red", "blue"), z) 
plot3d(x = x, y = y, z = z, col = cols)