getting 21 very different colours

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

getting 21 very different colours

Federico Calboli-3
Hi All,

I am plotting a scatterplot of 21 populations, and I am using rainbow(21)[pops.col] to generate 21 colours for the plot (which works).  Maybe it is because I can really process few colours at a time, but the differences between the colours are not as strong as I’d like.  I can specify start and end for rainbow(), but if anything that looks worse if I do not just stick to 0 and 1.  

Is there a way of getting a set of 21 colours that maximises the differences between them?  

I could pick them by hand, but that is about 15 colours more than I know (I have a detailed colourchart, but the visual differences between ’skyblue’ and ’slategrey’ elude me when plotted as dots on a plot).

Cheers

F
--
Federico Calboli
LBEG - Laboratory of Biodiversity and Evolutionary Genomics
Charles Deberiotstraat 32 box 2439
3000 Leuven
+32 16 32 87 67





______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: getting 21 very different colours

S Ellison-2
You could look at combning a number of palettes from the RColorBrewer package to get the palette length you want.

S Ellison

> -----Original Message-----
> From: R-help [mailto:[hidden email]] On Behalf Of Federico
> Calboli
> Sent: 11 September 2018 08:35
> To: [hidden email]
> Subject: [R] getting 21 very different colours
>
> Hi All,
>
> I am plotting a scatterplot of 21 populations, and I am using
> rainbow(21)[pops.col] to generate 21 colours for the plot (which works).
> Maybe it is because I can really process few colours at a time, but the
> differences between the colours are not as strong as I’d like.  I can specify
> start and end for rainbow(), but if anything that looks worse if I do not just
> stick to 0 and 1.
>
> Is there a way of getting a set of 21 colours that maximises the differences
> between them?
>
> I could pick them by hand, but that is about 15 colours more than I know (I
> have a detailed colourchart, but the visual differences between ’skyblue’ and
> ’slategrey’ elude me when plotted as dots on a plot).
>
> Cheers
>
> F
> --
> Federico Calboli
> LBEG - Laboratory of Biodiversity and Evolutionary Genomics
> Charles Deberiotstraat 32 box 2439
> 3000 Leuven
> +32 16 32 87 67
>
>
>
>
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.


*******************************************************************
This email and any attachments are confidential. Any use, copying or
disclosure other than by the intended recipient is unauthorised. If
you have received this message in error, please notify the sender
immediately via +44(0)20 8943 7000 or notify [hidden email]
and delete this message and any copies from your computer and network.
LGC Limited. Registered in England 2991879.
Registered office: Queens Road, Teddington, Middlesex, TW11 0LY, UK
______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: getting 21 very different colours

Ivan Krylov
In reply to this post by Federico Calboli-3
On Tue, 11 Sep 2018 07:34:51 +0000
Federico Calboli <[hidden email]> wrote:

> Is there a way of getting a set of 21 colours that maximises the
> differences between them?  

In my limited experience, getting even 10 colours to look different
enough is a serious undertaking. Take a look at RColorBrewer:
display.brewer.all(n, "qual") stops offering palettes for n>12.

When I needed a 10-colour categorical/qualitative palette, I opted for
brute force approach of maximising the minimal distance between points
in HCL colourspace, although later my colleague told me that I just
needed an existing algorithm to place the points uniformly. It has to
be HCL and not RGB because HCL signifies the way people perceive
different colours while RGB is only a good representation hardware-wise.

Here is my code; the usual disclaimers about stuff written between 1 and
3 AM apply:

# -------------------------8<---------------------------

require(nloptr)

h <- c(0,360)
c <- c(0,137) # see the warning about fixup in `?hcl`: not all HCL points are representable in RGB

# NOTE: depending on your plot background, you may have to change at least luminance range
l <- c(30,90)

npoints <- 24 # I had only 10 here

pts <- matrix(ncol=3, nrow=npoints, dimnames=list(NULL, c("h","c","l")))
pts[,"h"] <- runif(npoints, min=h[1], max=h[2])
pts[,"c"] <- runif(npoints, min=c[1], max=c[2])
pts[,"l"] <- runif(npoints, min=l[1], max=l[2])

lb <- cbind(h=rep(h[1],npoints), c=rep(c[1],npoints), l=rep(l[1],npoints))
ub <- cbind(h=rep(h[2],npoints), c=rep(c[2],npoints), l=rep(l[2],npoints))

obj <- function(x) {
        pts[,c("h","c","l")] <- x
        # somehow the best results were achieved by calculating Euclidean distance from cylindrical coordinates
        pts <- cbind(pts[,"c"]*sin(pts[,'h']/360*2*pi), pts[,'c']*cos(pts[,'h']/360*2*pi), pts[,'l'])
        d <- as.matrix(dist(pts))
        diag(d) <- NA
        # maximise minimal distance <=> minimize negative of minimal distance
        -min(d, na.rm=T)
}

# the stopping criterion is a bit lame, but the objective function here is very hard to minimize
# 1e6 iterations take a few minutes on a relatively modern desktop
sol <- nloptr(as.vector(pts), obj, lb=as.vector(lb), ub=as.vector(ub), opts=list(algorithm="NLOPT_GN_CRS2_LM", maxeval=1e6))

pts[,c("h","c",'l')] <- sol$solution

plot(pts[,"c"] * sin(pts[,"h"]/360*2*pi), pts[,"c"] * cos(pts[,"h"]/360*2*pi), col=hcl(pts[,"h"], pts[,"c"], l), pch=19, cex=2)

# -------------------------8<---------------------------

I couldn't get my code to produce 24 acceptably different colours, but
maybe you will succeed with a similar approach.

--
Best regards,
Ivan

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: getting 21 very different colours

Duncan Murdoch-2
In reply to this post by Federico Calboli-3
On 11/09/2018 3:34 AM, Federico Calboli wrote:
> Hi All,
>
> I am plotting a scatterplot of 21 populations, and I am using rainbow(21)[pops.col] to generate 21 colours for the plot (which works).  Maybe it is because I can really process few colours at a time, but the differences between the colours are not as strong as I’d like.  I can specify start and end for rainbow(), but if anything that looks worse if I do not just stick to 0 and 1.
>
> Is there a way of getting a set of 21 colours that maximises the differences between them?

The LAB and LUV color spaces (in the colorspace package) attempt to map
perceptual differences to equal distances.  You could try using a grid
of points in one of those spaces, but not all triples are valid.

However, 21 colours is probably too many for your purpose.  If you
really want to distinguish 21 groups, you're likely going to have to use
other characteristics as well, such as the symbol.  You could plot 21
different letters in 5 different colours and it might work, but it's not
going to be easy for viewers.

Duncan Murdoch

>
> I could pick them by hand, but that is about 15 colours more than I know (I have a detailed colourchart, but the visual differences between ’skyblue’ and ’slategrey’ elude me when plotted as dots on a plot).
>
> Cheers
>
> F
> --
> Federico Calboli
> LBEG - Laboratory of Biodiversity and Evolutionary Genomics
> Charles Deberiotstraat 32 box 2439
> 3000 Leuven
> +32 16 32 87 67
>
>
>
>
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: getting 21 very different colours

barry rowlingson
In reply to this post by Federico Calboli-3
On Tue, Sep 11, 2018 at 11:43 AM, Duncan Murdoch <[hidden email]>
wrote:

> On 11/09/2018 3:34 AM, Federico Calboli wrote:
> > Hi All,
> >
> > I am plotting a scatterplot of 21 populations, and I am using
> rainbow(21)[pops.col] to generate 21 colours for the plot (which works).
> Maybe it is because I can really process few colours at a time, but the
> differences between the colours are not as strong as I’d like.  I can
> specify start and end for rainbow(), but if anything that looks worse if I
> do not just stick to 0 and 1.
> >
> > Is there a way of getting a set of 21 colours that maximises the
> differences between them?
>
> The LAB and LUV color spaces (in the colorspace package) attempt to map
> perceptual differences to equal distances.  You could try using a grid
> of points in one of those spaces, but not all triples are valid.
>
> However, 21 colours is probably too many for your purpose.  If you
> really want to distinguish 21 groups, you're likely going to have to use
> other characteristics as well, such as the symbol.  You could plot 21
> different letters in 5 different colours and it might work, but it's not
> going to be easy for viewers.
>
>
The `alphabet` and `alphabet2` palettes from the `pals` package claim 26
"distinguishable" colours:

Details:

     The ‘alphabet’ palette has 26 distinguishable colors that have
     logical names starting with the English alphabet letters A, B, ...
     Z. This palette is based on the work by Green-Armytage (2010), but
     uses the names 'orange' instead of 'orpiment', and 'magenta'
     instead of 'mallow'.

There are some other palettes in that help page (?alphabet) that might also
work. But 21 colours is pushing it.

Barry






> Duncan Murdoch
>
> >
> > I could pick them by hand, but that is about 15 colours more than I know
> (I have a detailed colourchart, but the visual differences between
> ’skyblue’ and ’slategrey’ elude me when plotted as dots on a plot).
> >
> > Cheers
> >
> > F
> > --
> > Federico Calboli
> > LBEG - Laboratory of Biodiversity and Evolutionary Genomics
> > Charles Deberiotstraat 32 box 2439
> > 3000 Leuven
> > +32 16 32 87 67
> >
> >
> >
> >
> >
> > ______________________________________________
> > [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide http://www.R-project.org/
> posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
> >
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/
> posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: [FORGED] Re: getting 21 very different colours

Paul Murrell-2

You could also take a look at the 'Polychrome' package

Paul

On 12/09/18 03:36, Barry Rowlingson wrote:

> On Tue, Sep 11, 2018 at 11:43 AM, Duncan Murdoch <[hidden email]>
> wrote:
>
>> On 11/09/2018 3:34 AM, Federico Calboli wrote:
>>> Hi All,
>>>
>>> I am plotting a scatterplot of 21 populations, and I am using
>> rainbow(21)[pops.col] to generate 21 colours for the plot (which works).
>> Maybe it is because I can really process few colours at a time, but the
>> differences between the colours are not as strong as I’d like.  I can
>> specify start and end for rainbow(), but if anything that looks worse if I
>> do not just stick to 0 and 1.
>>>
>>> Is there a way of getting a set of 21 colours that maximises the
>> differences between them?
>>
>> The LAB and LUV color spaces (in the colorspace package) attempt to map
>> perceptual differences to equal distances.  You could try using a grid
>> of points in one of those spaces, but not all triples are valid.
>>
>> However, 21 colours is probably too many for your purpose.  If you
>> really want to distinguish 21 groups, you're likely going to have to use
>> other characteristics as well, such as the symbol.  You could plot 21
>> different letters in 5 different colours and it might work, but it's not
>> going to be easy for viewers.
>>
>>
> The `alphabet` and `alphabet2` palettes from the `pals` package claim 26
> "distinguishable" colours:
>
> Details:
>
>       The ‘alphabet’ palette has 26 distinguishable colors that have
>       logical names starting with the English alphabet letters A, B, ...
>       Z. This palette is based on the work by Green-Armytage (2010), but
>       uses the names 'orange' instead of 'orpiment', and 'magenta'
>       instead of 'mallow'.
>
> There are some other palettes in that help page (?alphabet) that might also
> work. But 21 colours is pushing it.
>
> Barry
>
>
>
>
>
>
>> Duncan Murdoch
>>
>>>
>>> I could pick them by hand, but that is about 15 colours more than I know
>> (I have a detailed colourchart, but the visual differences between
>> ’skyblue’ and ’slategrey’ elude me when plotted as dots on a plot).
>>>
>>> Cheers
>>>
>>> F
>>> --
>>> Federico Calboli
>>> LBEG - Laboratory of Biodiversity and Evolutionary Genomics
>>> Charles Deberiotstraat 32 box 2439
>>> 3000 Leuven
>>> +32 16 32 87 67
>>>
>>>
>>>
>>>
>>>
>>> ______________________________________________
>>> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide http://www.R-project.org/
>> posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>>>
>>
>> ______________________________________________
>> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/
>> posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

--
Dr Paul Murrell
Department of Statistics
The University of Auckland
Private Bag 92019
Auckland
New Zealand
64 9 3737599 x85392
[hidden email]
http://www.stat.auckland.ac.nz/~paul/

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: getting 21 very different colours

Zach Simpson
In reply to this post by Federico Calboli-3
Hi Federico

For a possible alternative, the scico package provides a nice
collection of color palettes that are designed to be both color-blind
friendly and differentiable:

https://www.data-imaginist.com/2018/scico-and-the-colour-conundrum/

You could generate a vector of 21 colors (spaced as far apart as
possible on the palette) to pass to your plot arguments with something
like:

library(scico)
scico(21, palette = 'oleron')

Not sure if this works for your case though. But maybe another feature
(shape?) could help differentiate the 21 points.

Hope this helps,
Zach Simpson

> Message: 11
> Date: Tue, 11 Sep 2018 07:34:51 +0000
> From: Federico Calboli <[hidden email]>
> To: "[hidden email]" <[hidden email]>
> Subject: [R] getting 21 very different colours
> Message-ID: <[hidden email]>
> Content-Type: text/plain; charset="utf-8"
>
> Hi All,
>
> I am plotting a scatterplot of 21 populations, and I am using rainbow(21)[pops.col] to generate 21 colours for the plot (which works).  Maybe it is because I can really process few colours at a time, but the differences between the colours are not as strong as I’d like.  I can specify start and end for rainbow(), but if anything that looks worse if I do not just stick to 0 and 1.
>
> Is there a way of getting a set of 21 colours that maximises the differences between them?
>
> I could pick them by hand, but that is about 15 colours more than I know (I have a detailed colourchart, but the visual differences between ’skyblue’ and ’slategrey’ elude me when plotted as dots on a plot).
>
> Cheers
>
> F
> --
> Federico Calboli
> LBEG - Laboratory of Biodiversity and Evolutionary Genomics
> Charles Deberiotstraat 32 box 2439
> 3000 Leuven
> +32 16 32 87 67

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: getting 21 very different colours

Achim Zeileis-4
Have a look at the Polychrome package by Kevin Coombes and Guy Brock:
https://CRAN.R-project.org/package=Polychrome

This employs the LUV space (with HCL = polar LUV) to get many distinct
distinguishable colors. For a few first steps, see:
https://CRAN.R-project.org/web/packages/Polychrome/vignettes/polychrome.html

On Wed, 12 Sep 2018, Zach Simpson wrote:

> Hi Federico
>
> For a possible alternative, the scico package provides a nice
> collection of color palettes that are designed to be both color-blind
> friendly and differentiable:
>
> https://www.data-imaginist.com/2018/scico-and-the-colour-conundrum/
>
> You could generate a vector of 21 colors (spaced as far apart as
> possible on the palette) to pass to your plot arguments with something
> like:
>
> library(scico)
> scico(21, palette = 'oleron')
>
> Not sure if this works for your case though. But maybe another feature
> (shape?) could help differentiate the 21 points.
>
> Hope this helps,
> Zach Simpson
>
>> Message: 11
>> Date: Tue, 11 Sep 2018 07:34:51 +0000
>> From: Federico Calboli <[hidden email]>
>> To: "[hidden email]" <[hidden email]>
>> Subject: [R] getting 21 very different colours
>> Message-ID: <[hidden email]>
>> Content-Type: text/plain; charset="utf-8"
>>
>> Hi All,
>>
>> I am plotting a scatterplot of 21 populations, and I am using rainbow(21)[pops.col] to generate 21 colours for the plot (which works).  Maybe it is because I can really process few colours at a time, but the differences between the colours are not as strong as I’d like.  I can specify start and end for rainbow(), but if anything that looks worse if I do not just stick to 0 and 1.
>>
>> Is there a way of getting a set of 21 colours that maximises the differences between them?
>>
>> I could pick them by hand, but that is about 15 colours more than I know (I have a detailed colourchart, but the visual differences between ’skyblue’ and ’slategrey’ elude me when plotted as dots on a plot).
>>
>> Cheers
>>
>> F
>> --
>> Federico Calboli
>> LBEG - Laboratory of Biodiversity and Evolutionary Genomics
>> Charles Deberiotstraat 32 box 2439
>> 3000 Leuven
>> +32 16 32 87 67
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.