using CairoMakie
using DataFrames
using Statistics
using RandomRadar Plot (Julia)
A radar (spider) plot profiles one or few samples across many comparable axes on polar coordinates — useful when each spoke is a score on the same 0–1 scale (immune signatures, facility QC panels), and misleading when scales differ. CairoMakie ships a PolarAxis block, but hand-projecting the spokes onto a cartesian axis keeps the outside labels fully under your control; the trigonometry is two lines.
Example
Setup
- System Requirements: Cross-platform (Linux/MacOS/Windows)
- Programming Language: Julia
- Dependencies:
CairoMakie,DataFrames,Statistics
Data Preparation
We score two CAR-T products on eight normalized (0–1) immune-function axes and convert them to spoke angles, closing the polygon by repeating the first point.
axes_names = ["Killing", "Cytokine", "Prolif.", "Persist.",
"Traffic", "Exhaust.", "Memory", "Safety"]
scores = DataFrame(
Product = ["CAR-19", "CAR-22"],
Values = [[0.9, 0.7, 0.8, 0.6, 0.5, 0.8, 0.65, 0.75],
[0.7, 0.85, 0.6, 0.8, 0.7, 0.6, 0.8, 0.85]])
K = length(axes_names)
angles = [2π * k / K for k in 0:K] # first angle repeated
profile(v) = [v; v[1]]
xy(r, θ) = Point2f(r * cos(θ), r * sin(θ))xy (generic function with 1 method)Visualization
Grid, Spokes, and a Single-Product Profile
Concentric guide rings, radial spokes, and axis labels are three small loops; the profile itself is one closed polygon.
θ_grid = range(0, 2π; length = 181)
fig = Figure(size = (480, 440))
ax = Axis(fig[1, 1]; aspect = DataAspect())
for r in 0.25:0.25:1.0 # guide rings
lines!(ax, [xy(r, θ) for θ in θ_grid]; color = :gray85, linewidth = 0.8)
end
for k in 1:K # spokes + labels
lines!(ax, [xy(0.0, angles[k]), xy(1.0, angles[k])];
color = :gray85, linewidth = 0.8)
text!(ax, xy(1.16, angles[k]); text = axes_names[k],
align = (:center, :center), fontsize = 11)
end
poly!(ax, [xy(v, θ) for (v, θ) in zip(profile(scores.Values[1]), angles)];
color = (:steelblue, 0.30), strokecolor = :steelblue, strokewidth = 1.8)
ax.xticks = []; ax.yticks = []
hidespines!(ax)
fig
Two-Product Overlay
Two semi-transparent fills on the same spokes make trade-offs (killing vs persistence) immediately visible.
θ_grid = range(0, 2π; length = 181)
fig = Figure(size = (540, 440))
ax = Axis(fig[1, 1]; aspect = DataAspect())
for r in 0.25:0.25:1.0
lines!(ax, [xy(r, θ) for θ in θ_grid]; color = :gray85, linewidth = 0.8)
end
for k in 1:K
lines!(ax, [xy(0.0, angles[k]), xy(1.0, angles[k])];
color = :gray85, linewidth = 0.8)
text!(ax, xy(1.16, angles[k]); text = axes_names[k],
align = (:center, :center), fontsize = 11)
end
product_colors = [:steelblue, :tomato]
for (i, row) in enumerate(eachrow(scores))
poly!(ax, [xy(v, θ) for (v, θ) in zip(profile(row.Values), angles)];
color = (product_colors[i], 0.25),
strokecolor = product_colors[i], strokewidth = 1.8)
end
elements = [PolyElement(polycolor = c) for c in product_colors]
Legend(fig[1, 2], elements, scores.Product, framevisible = false)
ax.xticks = []; ax.yticks = []
hidespines!(ax)
fig
References
- Makie development team. Makie.jl documentation — PolarAxis. https://docs.makie.org/stable/reference/blocks/polaraxis
- Danisch, S., & Krumbiegel, J. (2021). Makie.jl: Flexible high-performance data visualization for Julia. Journal of Open Source Software, 6(65), 3349.