Various data in the upper and lower panels of the diffused screen matrix

Question

I want to build two different datasets in a scatter plot .

I know that I can use upper.panel and lower.panel to differentiate the plot function. However, I am unable to put my data in a suitable format to use this.

Suppose I have two tissues (“brain” and “heart”) and four conditions (1-4). Now I can use, for example. pairs(data$heart) to get the scatter matrix for one of the data sets. Suppose I have the following data:

 conditions <- 1 : 4 noise <- rnorm(100) data <- list(brain = sapply(conditions, function (x) noise + 0.1 * rnorm(100)), heart = sapply(conditions, function (x) noise + 0.3 * rnorm(100))) 

How to do this in a format so that pairs(data, …) display one data set above and one below the diagonal, as shown here (green = brain, purple = heart):

Just using

 pairs(data, upper.panel = something, lower.panel = somethingElse) 

It does not work, because it will display all conditions depending on all conditions without taking into account different fabrics - it, in fact, ignores the list when reordering the hierarchy (i.e., it has data = (A=list(brain=…, heart=…), B=list(brain=…, heart=…), …) ).

Answer 1

This is the best I can do using passing arguments:

 foo.upper <- function(x,y,ind.upper,col.upper,ind.lower,col.lower,...){ points(x[ind.upper],y[ind.upper],col = col.upper,...) } foo.lower <- function(x,y,ind.lower,col.lower,ind.upper,col.upper,...){ points(x[ind.lower],y[ind.lower],col = col.lower,...) } pairs(dat[,-5], lower.panel = foo.lower, upper.panel = foo.upper, ind.upper = dat$type == 'brain', ind.lower = dat$type == 'heart', col.upper = 'blue', col.lower = 'red') 

Note that each panel requires all arguments. ... is a cruel lover. If you include only specific panel arguments in each function, it works, but you get many, many warnings from R trying to pass these arguments to regular build functions and obviously they won't.

This was my first attempt, but it seems ugly:

 dat <- as.data.frame(do.call(rbind,data)) dat$type <- rep(c('brain','heart'),each = 100) foo.upper <- function(x,y,...){ points(x[dat$type == 'brain'],y[dat$type == 'brain'],col = 'red',...) } foo.lower <- function(x,y,...){ points(x[dat$type == 'heart'],y[dat$type == 'heart'],col = 'blue',...) } pairs(dat[,-5],lower.panel = foo.lower,upper.panel = foo.upper) 

I am abusing R, considering the somewhat ugly path here in this second version. (Of course, you could do it more cleanly in a grid , but you probably knew that.)

The only other option I can think of is to create my own matrix of scattering matrices using layout , but this is probably quite a bit of work.

Lattice Editing

Here, at least, is the beginning of a lattice solution. It should handle different ranges of x, y axes better, but I have not tested this.

 dat <- do.call(rbind,data) dat <- as.data.frame(dat) dat$grp <- rep(letters[1:2],each = 100) plower <- function(x,y,grp,...){ panel.xyplot(x[grp == 'a'],y[grp == 'a'],col = 'red',...) } pupper <- function(x,y,grp,...){ panel.xyplot(x[grp == 'b'],y[grp == 'b'],...) } splom(~dat[,1:4], data = dat, lower.panel = plower, upper.panel = pupper, grp = dat$grp)