plot.la: Visualizing the dynamic correlation between a pair of genes...
In DCA: Dynamic Correlation Analysis for High Dimensional Data

Description Usage Arguments Details Value Author(s) Examples

Given three vectors, x, y and z, the function produces a color scatter plot of x and y, colored by the grouping of z.

1	plot_la(x, y, z, use.locfdr = FALSE, cols = c("red", "green", "blue"), cex = 0.5)

`x`	The x vector.
`y`	The y vector.
`z`	The z vector, conditioned on which X and Y have dynamic correlation.
`use.locfdr`	Whether to use local fdr to group the z values. If TRUE, locfdr() is used and the fdr (posterior probability of belonging to the more extreme groups) threshold is 0.5. If FALSE, the z values are cut at the 0.33 and 0.67 quantiles into three groups.
`cols`	The colors of the groups.
`cex`	The point size to be passed to plot().

The locfdr approach is only to be used when it is clear z has heavy tails, and it is believed such points are driving the dynamic correlation.

A vector, each item is a string, recording the LA score and the correlation in each of the point groups as determined by z values.

Tianwei Yu <tianwei.yu@emory.edu>

1 2	x<-la.simu.gen(n=100,p=20,n.grp=2, n.noise.gene=10, rho=0.25, pwr=1) plot_la(x$dat[1,], x$dat[2,], x$z[,1],use.locfdr=FALSE)

[1] "LA score:  45.9"                        
[2] "group: 1 ( -0.74 ) red , corr= -0.668"  
[3] "group: 2 ( 0.0032 ) green , corr= 0.355"
[4] "group: 3 ( 1.1 ) blue , corr= 0.643"