gl.grm.network 'names' attribute must be same length as vector

[Matthew Holen]

Hello, 

I'm attempting to use the gl.grm.network function and continue to get an error of 
Error in names(colors_pops) <- as.character(levels(x$pop)) : 'names' attribute [96] must be the same length as the vector [11] In addition: Warning message: In RColorBrewer::brewer.pal(ncolors, palette) : n too large, allowed maximum for palette PuOr is 11 Returning the palette you asked for with that many colors

with the default as:

palette_discrete = gl.select.colors(x, library = "brewer", palette = "PuOr", ncolors = nPop(x), verbose = 0)

Whenever I plot a PCA it correctly assigns colors (albeit not amazing colors), how should I go about getting gl.grm.network to do something similar?

The large number of populations is the use of accessions rather than species or genus as labels. 

Thank you!

[Bernd.Gruber]

Hi  Matthew,

 

The brewer pals allows only max 11 colors, and you seem to need 96 colors (as it seems). Hence you need to find another way to create your colors using a different library.

 

@Luis.Mijangos Araujo: not sure if the brewer library is hard coded in gl.grm.network, if so this is an error on our side.

 

Cheers, Bernd

--
You received this message because you are subscribed to the Google Groups "dartR" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dartr+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/dartr/178e955f-739e-426e-83ef-d00acadbc7ebn%40googlegroups.com.

[Jose Luis Mijangos]

Hi,

Please give a try to the below code,

Cheers,

Luis 

# Load core packages
library(dartRverse)   # dartR tools and data structures for genetic analyses
library(colorspace)   # advanced color palettes (e.g., HCL-based)

# Compute genomic relationship matrix from a genlight object
t1 <- testset.gl
r1 <- gl.grm(t1)

# Number of populations (to determine how many colors we need)
n_cols <- nPop(t1)

# Custom “polychrome” palette: an arbitrary base set of 36 colors
# • darken() applies a slight darkening so colors remain distinct
# • rep(...)[1:n] repeats the palette infinitely, so you can request any n
polychrome <- function(n) {
  p <- darken(
    setNames(
      c(
        "#3283FE","#FEAF16","#B00068","#1CFFCE","#90AD1C","#2ED9FF",
        "#DEA0FD","#AA0DFE","#F8A19F","#325A9B","#C4451C","#1C8356",
        "#85660D","#B10DA1","#FBE426","#1CBE4F","#FA0087","#FC1CBF",
        "#F7E1A0","#C075A6","#782AB6","#AAF400","#BDCDFF","#822E1C",
        "#B5EFB5","#7ED7D1","#1C7F93","#D85FF7","#683B79","#66B0FF",
        "#5A5156","#E4E1E3","#F6222E","#FE00FA","#16FF32","#3B00FB"
      ),
      c(
        "Dark_Purplish_Gray","Purplish_White","Vivid_Red","Vivid_Purple",
        "Vivid_Yellowish_Green","Strong_Purplish_Blue","Vivid_Orange_Yellow",
        "Vivid_Purplish_Red","Brilliant_Green","Vivid_Yellow_Green",
        "Vivid_Blue","Brilliant_Purple","Vivid_Violet","Strong_Pink",
        "Strong_Blue","Strong_Reddish_Orange","Vivid_Green","Light_Olive_Brown",
        "Vivid_Reddish_Purple","Vivid_Greenish_Yellow","Vivid_Yellowish_Green",
        "Vivid_Red","Vivid_Purplish_Red","Pale_Yellow","Strong_Reddish_Purple",
        "Vivid_Violet","Vivid_Yellow_Green","Very_Light_Blue","Strong_Reddish_Brown",
        "Very_Light_Yellowish_Green","Very_Light_Bluish_Green","Deep_Greenish_Blue",
        "Vivid_Purple","Deep_Purple","Brilliant_Blue","Vivid_Violet"
      )
    ),
    amount = 0.2
  )
  # Return exactly n colors by recycling the base palette
  rep(p, ceiling(n/length(p)))[1:n]
}

# Palette 1: our custom “polychrome” – truly unlimited via recycling
pal1 <- polychrome(n_cols)

# Palette 2: HCL-based rainbow – smooth hues around the circle
# rainbow_hcl() can generate any number of distinct hues
pal2 <- rainbow_hcl(n_cols)

# Palette 3: Predefined HCL sequence (“Dark 3”) – unlimited length
# Good for qualitative displays with controlled contrast
pal3 <- hcl.colors(n_cols, palette = "Dark 3")

# Palette 4: Base-R rainbow via dartR helper – unlimited
# gl.select.colors() wraps base R’s palette functions
pal4 <- gl.select.colors(library = 'baseR', palette = 'rainbow', ncolors = n_cols)

# Build and plot a GRM network, coloring populations with an unlimited palette
r2 <- gl.grm.network(
  G = r1,
  x = t1,
  standardise = TRUE,
  palette_discrete = pal4  # you can swap in pal1, pal2, or pal3 as needed
)

[Matthew Holen]

Luis,

Thank you for the quick reply. The listed Palette 4 worked great. 

Best,

Matt