Add a custom border to certain cells in a Matplotlib / Seaborn plot

Adding custom borders to specific cells in Matplotlib/Seaborn plots helps highlight important data points. You can use Rectangle patches to create colored borders around target cells in heatmaps.

Basic Approach

The process involves creating a heatmap, accessing its axes, and adding rectangle patches with custom colors and line widths ?

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Create sample data
data = pd.DataFrame({
    "col1": [1, 4, 2, 3, 5], 
    "col2": [3, 4, 1, 5, 2]
})

# Create clustered heatmap
g = sns.clustermap(data, figsize=(7.5, 3.5))
ax = g.ax_heatmap

# Add custom border to specific cells
border_color = "yellow"
ax.add_patch(plt.Rectangle((1, 2), 2, 1, fill=False,
                          edgecolor=border_color, lw=5))

plt.show()

Multiple Custom Borders

You can add multiple borders with different colors and styles to highlight various cell groups ?

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Create sample dataset
data = pd.DataFrame({
    "A": [1, 4, 2, 3, 5], 
    "B": [3, 4, 1, 5, 2],
    "C": [2, 1, 4, 2, 3]
})

# Create heatmap
plt.figure(figsize=(8, 6))
ax = sns.heatmap(data, annot=True, cmap="Blues")

# Add multiple custom borders
# Red border around single cell
ax.add_patch(plt.Rectangle((0, 1), 1, 1, fill=False,
                          edgecolor="red", lw=3))

# Green border around 2x2 area
ax.add_patch(plt.Rectangle((1, 2), 2, 2, fill=False,
                          edgecolor="green", lw=4))

# Yellow dashed border
ax.add_patch(plt.Rectangle((0, 3), 3, 1, fill=False,
                          edgecolor="yellow", lw=2, linestyle="--"))

plt.title("Heatmap with Custom Cell Borders")
plt.show()

Rectangle Parameters

Dynamic Border Selection

You can programmatically add borders based on data conditions ?

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np

# Create sample data
data = pd.DataFrame(np.random.randint(1, 10, (4, 4)), 
                   columns=["A", "B", "C", "D"])

plt.figure(figsize=(8, 6))
ax = sns.heatmap(data, annot=True, cmap="viridis")

# Add borders to cells with values greater than 6
for i in range(data.shape[0]):
    for j in range(data.shape[1]):
        if data.iloc[i, j] > 6:
            ax.add_patch(plt.Rectangle((j, i), 1, 1, fill=False,
                                     edgecolor="red", lw=3))

plt.title("Borders Around High Values (> 6)")
plt.show()

Conclusion

Use Rectangle patches with add_patch() to create custom cell borders in heatmaps. Adjust position, size, color, and line style to highlight specific data patterns effectively.