Correlogram (Julia)

Authors

[Editor] Shixiang Wang;

[Contributors] Claude Code — AI-generated, independently AI-reviewed;

Modified

2026-10-19

🤖 AI Skill — Download the Bizard unified skill for your AI assistant ⬇️ Download Skill ZIP
🔌 MCP Server — Let AI agents (Claude Code, Cursor, …) query Bizard live: claude mcp add bizard -- npx @openbiox/bizard-mcp — or the free hosted instance (no install, faster from mainland China): claude mcp add --transport http bizard http://biotree.top:38123/bizard-mcp/mcp 📖 MCP Docs
📖 Cite Bizard — if Bizard helps your research, please cite: Li, K., Zheng, H., Huang, K., et al. (2026). Bizard: A Community-Driven Platform for Accelerating and Enhancing Biomedical Data Visualization. iMetaMed, e70038. 🔗 DOI / BibTeX

A correlogram (correlation matrix heatmap) encodes every pairwise correlation of a feature panel as a colored matrix cell, giving an instant read on co-expression blocks and redundant biomarkers. In biomedical research it is routine QC for marker panels. Julia’s Statistics.cor computes the matrix and CairoMakie.heatmap! renders it, with cell values annotated via text!.

Example

Setup

  • System Requirements: Cross-platform (Linux/MacOS/Windows)
  • Programming Language: Julia
  • Dependencies: CairoMakie, DataFrames, Statistics 
using CairoMakie
using DataFrames
using Statistics
using Random

Data Preparation

We simulate an 8-marker immune panel with two correlated blocks (T-cell markers and myeloid markers) plus a weakly related proliferation score, then compute the Pearson matrix.

Random.seed!(101)

n = 250
latent_t = randn(n)
latent_m = randn(n)
panel = DataFrame(
    CD3   = latent_t .+ 0.4 .* randn(n),
    CD4   = latent_t .+ 0.5 .* randn(n),
    CD8   = latent_t .+ 0.5 .* randn(n),
    IL7R  = latent_t .+ 0.6 .* randn(n),
    CD14  = latent_m .+ 0.4 .* randn(n),
    CD68  = latent_m .+ 0.5 .* randn(n),
    LYZ   = latent_m .+ 0.5 .* randn(n),
    MKI67 = randn(n))
markers = names(panel)
corr = Matrix(cor(Matrix(panel)))
8×8 Matrix{Float64}:
  1.0         0.847918    0.851106   …  -0.0763281  -0.121005   -0.0607499
  0.847918    1.0         0.830831      -0.0535264  -0.0561228  -0.0197228
  0.851106    0.830831    1.0           -0.0657583  -0.0852999  -0.0260574
  0.817423    0.795928    0.77068       -0.0428039  -0.0724778  -0.129194
 -0.0678254  -0.0418892  -0.0612808      0.835868    0.832141   -0.00942036
 -0.0763281  -0.0535264  -0.0657583  …   1.0         0.803521    0.0111169
 -0.121005   -0.0561228  -0.0852999      0.803521    1.0         0.0349363
 -0.0607499  -0.0197228  -0.0260574      0.0111169   0.0349363   1.0

Visualization

Annotated Correlation Heatmap

A diverging colormap reversed so red means positive, a fixed colorrange of ±1 for cross-study comparability, and one text! per cell keep the numbers readable.

fig = Figure(size = (540, 440))
ax = Axis(fig[1, 1]; xticks = (1:8, markers), yticks = (1:8, markers),
          xticklabelrotation = 0.6)
hm = heatmap!(ax, corr; colormap = Reverse(:RdBu), colorrange = (-1, 1))
for i in 1:8, j in 1:8
    text!(ax, j, i; text = string(round(corr[i, j]; digits = 2)),
          align = (:center, :center), fontsize = 9,
          color = abs(corr[i, j]) > 0.55 ? :white : :black)
end
Colorbar(fig[1, 2], hm; label = "Pearson r")
fig
Figure 1: Marker-panel correlogram with annotated coefficients

Lower-Triangle Correlogram

Masking the mirrored upper triangle with NaN removes the redundancy; nan_color keeps the empty cells invisible.

corr_masked = [i < j ? NaN : corr[i, j] for i in 1:8, j in 1:8]

fig = Figure(size = (460, 420))
ax = Axis(fig[1, 1]; xticks = (1:8, markers), yticks = (1:8, markers),
          xticklabelrotation = 0.6)
heatmap!(ax, corr_masked; colormap = Reverse(:RdBu), colorrange = (-1, 1),
         nan_color = :white)
for i in 1:8, j in 1:i
    text!(ax, j, i; text = string(round(corr[i, j]; digits = 2)),
          align = (:center, :center), fontsize = 9,
          color = abs(corr[i, j]) > 0.55 ? :white : :black)
end
hidespines!(ax)
fig
Figure 2: Lower-triangle correlogram (upper triangle masked)

References

  1. Makie development team. Makie.jl documentation — heatmap. https://docs.makie.org/stable/reference/plots/heatmap
  2. Danisch, S., & Krumbiegel, J. (2021). Makie.jl: Flexible high-performance data visualization for Julia. Journal of Open Source Software, 6(65), 3349.