Bio Data Visualization Color Palettes

Version Compatibility

Reference examples tested with: ggplot2 3.5+, matplotlib 3.8+, seaborn 0.13+

Before using code patterns, verify installed versions match. If versions differ:

• Python: pip show <package> then help(module.function) to check signatures
• R: packageVersion('<pkg>') then ?function_name to verify parameters

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

Color Palettes

"Choose colors for a biological visualization" → Select appropriate color palettes for sequential data (expression levels), diverging data (fold changes), or categorical data (cell types/clusters).

• Python: seaborn.color_palette(), matplotlib.colormaps
• R: RColorBrewer::brewer.pal(), viridis::viridis()

Palette Types

viridis (Colorblind-Safe)

library(viridis)

# Continuous scale
ggplot(df, aes(x, y, color = value)) +
    geom_point() +
    scale_color_viridis_c()

# Discrete scale
ggplot(df, aes(x, y, color = group)) +
    geom_point() +
    scale_color_viridis_d()

# Options: viridis, magma, plasma, inferno, cividis, turbo
scale_color_viridis_c(option = 'magma')

import matplotlib.pyplot as plt

plt.scatter(x, y, c=values, cmap='viridis')
# Options: viridis, magma, plasma, inferno, cividis

RColorBrewer (R)

library(RColorBrewer)

# View all palettes
display.brewer.all()

# Sequential
scale_fill_brewer(palette = 'Blues')
scale_color_distiller(palette = 'YlOrRd', direction = 1)

# Diverging
scale_fill_brewer(palette = 'RdBu')
scale_color_gradient2(low = '#4DBBD5', mid = 'white', high = '#E64B35', midpoint = 0)

# Qualitative
scale_color_brewer(palette = 'Set1')
scale_fill_brewer(palette = 'Dark2')

# Get colors directly
brewer.pal(n = 5, name = 'Set1')

matplotlib/seaborn (Python)

import matplotlib.pyplot as plt
import seaborn as sns

# Sequential
plt.scatter(x, y, c=values, cmap='Blues')

# Diverging
plt.scatter(x, y, c=values, cmap='RdBu_r', vmin=-2, vmax=2)

# Qualitative
palette = sns.color_palette('Set1', n_colors=5)
sns.scatterplot(x=x, y=y, hue=group, palette='Set1')

# Custom palette
custom_palette = {'Control': '#4DBBD5', 'Treatment': '#E64B35'}
sns.scatterplot(x=x, y=y, hue=group, palette=custom_palette)

Scientific Journal Palettes

library(ggsci)

# Nature Publishing Group
scale_color_npg()
scale_fill_npg()

# AAAS Science
scale_color_aaas()

# Lancet
scale_color_lancet()

# JAMA
scale_color_jama()

# JCO
scale_color_jco()

Custom Palettes

# Define custom colors
my_colors <- c(
    'Control' = '#4DBBD5',
    'Treatment' = '#E64B35',
    'Vehicle' = '#00A087'
)

scale_color_manual(values = my_colors)
scale_fill_manual(values = my_colors)

# Create gradient
colorRampPalette(c('blue', 'white', 'red'))(100)

from matplotlib.colors import LinearSegmentedColormap

colors = ['#4DBBD5', 'white', '#E64B35']
cmap = LinearSegmentedColormap.from_list('custom_diverging', colors)
plt.imshow(data, cmap=cmap)

Heatmap Colors

library(circlize)

# For ComplexHeatmap
col_fun <- colorRamp2(c(-2, 0, 2), c('#4DBBD5', 'white', '#E64B35'))

# For pheatmap
pheatmap(mat, color = colorRampPalette(rev(brewer.pal(9, 'RdBu')))(100))

import seaborn as sns

sns.heatmap(data, cmap='RdBu_r', center=0, vmin=-2, vmax=2)

Colorblind Simulation

Goal: Verify that a chosen palette remains distinguishable under common forms of color vision deficiency.

Approach: Use the colorspace package to simulate deuteranopia and protanopia transformations on the palette colors and visually inspect the result.

library(colorspace)

# Check if palette is colorblind safe
demoplot(rainbow(5), type = 'map')
demoplot(viridis(5), type = 'map')

# Simulate colorblindness
cvd_colors <- deutan(c('#E64B35', '#4DBBD5', '#00A087'))  # deuteranopia
cvd_colors <- protan(c('#E64B35', '#4DBBD5', '#00A087'))  # protanopia

Recommended Palettes

Transparency

# Add alpha
scale_color_manual(values = alpha(c('#E64B35', '#4DBBD5'), 0.7))

# In geom
geom_point(alpha = 0.6)

# Add alpha to hex
def add_alpha(hex_color, alpha):
    return hex_color + format(int(alpha * 255), '02x')

color_with_alpha = add_alpha('#E64B35', 0.7)

# In scatter
plt.scatter(x, y, c='#E64B35', alpha=0.7)

Extract Colors from Palette

# Get discrete colors
pal <- brewer.pal(8, 'Set1')
pal[1:3]  # First 3 colors

# Interpolate more colors
colorRampPalette(brewer.pal(8, 'Set1'))(20)

import seaborn as sns

palette = sns.color_palette('Set1', n_colors=8)
palette[:3]  # First 3 colors

# As hex
palette.as_hex()

Related Skills

• data-visualization/ggplot2-fundamentals - Apply colors
• data-visualization/heatmaps-clustering - Heatmap colors
• data-visualization/specialized-omics-plots - Plot styling