CODEWITHSUNDEEP
rcolorsplotrectanglescolor-palette
kurdtc
kurdtc

Reputation: 1621

R: Assign colors to values/color gradient palette

I have a sample dataframe which looks like this:

reg1 <- structure(list(REGION = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L,  1L, 1L, 2L, 2L, 2L, 2L, 2L), .Label = c("REG1", "REG2"), class = "factor"),STARTYEAR = c(1959L, 1960L, 1961L, 1962L, 1963L, 1964L, 1965L, 1966L, 1967L, 1945L, 1946L, 1947L, 1948L, 1949L), ENDYEAR = c(1960L, 1961L, 1962L, 1963L, 1964L, 1965L, 1966L, 1967L, 1968L, 1946L, 1947L, 1948L, 1949L, 1950L), Y_START = c(0L, 0L, 0L, 0L,  0L, 0L, 0L, 0L, 0L, 2L, 2L, 2L, 2L, 2L), Y_END = c(1L, 1L,  1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 3L, 3L, 3L, 3L), COLOR_VALUE = c(-969L, -712L, -574L, -312L, -12L, 1L, 0L, -782L, -999L, -100L, 23L, 45L, NA, 999L)), .Names = c("REGION", "STARTYEAR", "ENDYEAR", "Y_START", "Y_END", "COLOR_VALUE"), class = "data.frame", row.names = c(NA, -14L))

   REGION STARTYEAR ENDYEAR Y_START Y_END COLOR_VALUE
1    REG1      1959    1960       0     1        -969
2    REG1      1960    1961       0     1        -712
3    REG1      1961    1962       0     1        -574
4    REG1      1962    1963       0     1        -312
5    REG1      1963    1964       0     1         -12
6    REG1      1964    1965       0     1           1
7    REG1      1965    1966       0     1           0
8    REG1      1966    1967       0     1        -782
9    REG1      1967    1968       0     1        -999
10   REG2      1945    1946       2     3        -100
11   REG2      1946    1947       2     3          23
12   REG2      1947    1948       2     3          45
13   REG2      1948    1949       2     3          NA
14   REG2      1949    1950       2     3         999

I am creating a plot with the rect() function which works fine.

xx = unlist(reg1[, c(2, 3)])
yy = unlist(reg1[, c(4, 5)])
png(width=1679, height=1165, res=150)
if(any(xx < 1946)) {my_x_lim <- c(min(xx), 2014)} else {my_x_lim <- c(1946, 2014)} 
plot(xx, yy, type='n', xlim = my_x_lim)
apply(reg1, 1, function(y) 
  rect(y[2], y[4], y[3], y[5]))
dev.off()

In my reg1 data I have a 6th column which contains values between +1000 and -1000. What I was wondering is if there is a method that I could colour the rectangles in my plot according to my color values. Low values should be blue, values around 0 should result in white and high values in red (if no value is present or NA, then grey should be plotted).

My question: How could I create a color palette that ranges from values 1000 to -1000 (from red over white to blue) and apply it to my plot so that each rectangle gets coloured according to the color value?

Upvotes: 0

Views: 4308

Answers (1)

ECII
ECII

Reputation: 10629

Here is how your get a color ramp and match it in the data frame. 

my.colors<-colorRampPalette(c("blue", "white", "red")) #creates a function my.colors which interpolates n colors between blue, white and red
color.df<-data.frame(COLOR_VALUE=seq(-1000,1000,1), color.name=my.colors(2001)) #generates 2001 colors from the color ramp
reg1.with.color<-merge(reg1, color.df, by="COLOR_VALUE")

I can't help you with the rect() plotting, I've never used it

Upvotes: 4

Related Questions