Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Using Conditional Statements to Change the Color of Data Points

I have a data set, which I have used to make a scatter plot and I would like to assign three different colors to the data points within three different regions, based on their x values.

Data points with x-values < 3 I want to appear red

Data points with x-values (3,1549) I want to appear black

Data points with x values >1549 I want to appear purple

Here is my code for the scatterplot and accomplishing the first two parameters, how might I implement the third parameter, so that the last region's data points will be purple?

plot(x, y, xlab="chr X position (Mb)",
           ylab="Diversity",
           pch=16, cex =0.7, 
           col = ifelse(x < 3,'red','black'))    

enter image description here

like image 960
martin gordy Avatar asked Oct 15 '13 07:10

martin gordy


4 Answers

Just use nested ifelses:

plot(...., col=ifelse(x < 3, "red", ifelse(x > 1549, "purple", "black")))
like image 135
Hong Ooi Avatar answered Nov 08 '22 22:11

Hong Ooi


Also, the "classic" findInterval:

col = c("red", "black", "purple")[findInterval(x, v = c(0,3,1549))]

like image 32
alexis_laz Avatar answered Nov 08 '22 22:11

alexis_laz


You can define a vector of colors and pass it to the col argument of plot. Something like this :

colors <- rep("black", length(x))
colors[x<3] <- "red"
colors[x>1549] <- "pink"

plot(x, y, xlab="chr X position (Mb)",
           ylab="Diversity",
           pch=16, cex =0.7, 
           col = colors)    
like image 23
juba Avatar answered Nov 08 '22 23:11

juba


I like the cut approach:

set.seed(1)
x <- sample(1600)
col <- c("red", "black", "purple")
col <- col[cut(x, breaks=c(-Inf, 3, 1549, Inf))]
like image 45
sgibb Avatar answered Nov 08 '22 23:11

sgibb