Multipanel Figures

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: no behavioural pass.

Audit 0.4.1 · source sha c8c1fc3a2b13 · 19 Oct 2026Report an issue or request a re-scan

• PublisherGPTomics GitHub organizationOwner of the repository this skill is read from
• Source folderGPTomics/bioSkills/data-visualization/multipanel-figuresThe exact folder we audited
• LicenceMITAs the repository states it
• Commit pinned for this verdictc8c1fc3a2b13A new audit runs when the source changes
• Repository statusArchivedLast push 65 days ago
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Compose multi-panel publication figures with patchwork, cowplot, gridExtra (R), or matplotlib GridSpec/subfigures (Python) including shared axes/legends/guides collection, panel labels in Nature/Cell convention, and…

---
name: bio-data-visualization-multipanel-figures
description: Compose multi-panel publication figures with patchwork, cowplot, gridExtra (R), or matplotlib GridSpec/subfigures (Python) including shared axes/legends/guides collection, panel labels in Nature/Cell convention, and journal-spec sizing. Covers patchwork ≥1.2.0 axes='collect' feature, Type-42 font embedding, and the cairo_pdf save path. Use when composing 2+ subpanels into a single figure for journal submission.
tool_type: mixed
primary_tool: patchwork
---

## Version Compatibility

Reference examples tested with: patchwork 1.2+ (axes='collect' requires this version, released 2024-01-05), cowplot 1.1+, ggplot2 3.5+, matplotlib 3.8+ (subfigures stable since 3.4).

Before using code patterns, verify installed versions match. If versions differ:
- R: `packageVersion('<pkg>')` then `?function_name`
- Python: `pip show <package>` then `help(module.function)`

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

# Multi-Panel Figures

**"Combine plots into a multi-panel figure"** -> Arrange individual plots into a single composed figure with consistent sizing, shared legends/axes, and panel labels (a, b, c) in the Nature/Cell convention. The decision space: which composition library (patchwork most modern in R; matplotlib subfigures in Python), how to share legends and axes, and how to size at journal specifications.

- R: `patchwork` (modern; supports axes/guides collection since 1.2), `cowplot` (older; align_plots), `gridExtra` (basic grid arrange)
- Python: `matplotlib.gridspec.GridSpec`, `fig.subfigures()` (matplotlib 3.4+)

## The Single Most Important Modern Insight -- Axes Collection Requires patchwork ≥ 1.2.0

patchwork 1.2.0 (released 2024-01-05) added `axes = 'collect'` and `axis_titles = 'collect'` to `plot_layout()`. These collect repeated axes / titles across subplots into a single shared axis label — the same way `guides = 'collect'` (available since patchwork 1.0) collects legends.

Without this, multi-panel figures with shared axes show redundant labels on every subplot (visually cluttered AND non-Nature compliant). Verify patchwork version is ≥ 1.2.0; older versions silently ignore the `axes` argument.

## patchwork -- Modern R Composition

**Goal:** Compose 4 ggplot objects into a 2×2 panel figure with shared legend, collected axes, and bold panel labels (a, b, c, d) in upper-left of each subplot.

**Approach:** Combine plots with `+`, `/`, `|` operators; apply `plot_layout(guides='collect', axes='collect')` for shared elements; add `plot_annotation(tag_levels='a')` for Nature-style panel labels.

```r
library(patchwork)
library(ggplot2)

p1 <- ggplot(df, aes(x, y)) + geom_point() + theme_classic()
p2 <- ggplot(df, aes(group, value)) + geom_boxplot() + theme_classic()
p3 <- ggplot(df, aes(x)) + geom_histogram() + theme_classic()
p4 <- ggplot(df, aes(x, y, color = group)) + geom_point() + theme_classic()

# 2x2 grid
fig <- (p1 + p2) / (p3 + p4) +
    plot_annotation(tag_levels = 'a',
                    theme = theme(plot.tag = element_text(face = 'bold', size = 10))) +
    plot_layout(guides = 'collect',         # share legends
                axes = 'collect',           # share axes (patchwork >= 1.2.0)
                axis_titles = 'collect')

ggsave('figure1.pdf', fig, width = 180, height = 140, units = 'mm', device = cairo_pdf)
```

## patchwork Operators

```r
p1 + p2                                     # side-by-side
p1 / p2                                     # vertical stack
(p1 | p2) / p3                              # mixed: top row two, bottom one
p1 + p2 + p3 + plot_layout(ncol = 3)
p1 + p2 + plot_layout(widths = c(2, 1))     # 2:1 width ratio

# Complex grid via design string
design <- "
AAB
AAB
CCC
"
p1 + p2 + p3 + plot_layout(design = design)

# Inset
p1 + inset_element(p2, left = 0.6, bottom = 0.6, right = 1, top = 1)
```

## cowplot -- Alternative with Alignment Focus

```r
library(cowplot)

# plot_grid is the workhorse
combined <- plot_grid(p1, p2, p3, p4,
                       ncol = 2, labels = 'AUTO',         # 'AUTO' = A, B, C, D
                       label_size = 12, label_fontface = 'bold',
                       align = 'hv',                       # align horizontally + vertically
                       rel_widths = c(1, 1), rel_heights = c(1, 1))

# Nested grids
top_row <- plot_grid(p1, p2, ncol = 2, labels = c('A', 'B'))
bottom <- plot_grid(p3, p4, ncol = 2, labels = c('C', 'D'))
combined <- plot_grid(top_row, bottom, nrow = 2, rel_heights = c(1, 1.2))

ggsave('figure.pdf', combined, width = 180, height = 140, units = 'mm', device = cairo_pdf)
```

cowplot is older but its alignment behavior is sometimes more reliable than patchwork on edge cases (axes-with-titles of different lengths).

## matplotlib GridSpec (Python)

```python
import matplotlib.pyplot as plt
from matplotlib.gridspec import GridSpec

fig = plt.figure(figsize=(180/25.4, 120/25.4), constrained_layout=True)
gs = GridSpec(2, 3, figure=fig)

ax1 = fig.add_subplot(gs[0, 0])
ax2 = fig.add_subplot(gs[0, 1:])         # top right, spans columns 1-2
ax3 = fig.add_subplot(gs[1, :])           # bottom row, spans all columns

ax1.scatter(x, y, s=4, rasterized=True)
ax2.plot(x, y)
ax3.bar(cats, vals)

# Panel labels at (-0.15, 1.05) of each axes
for ax, lbl in zip([ax1, ax2, ax3], 'abc'):
    ax.text(-0.15, 1.05, lbl, transform=ax.transAxes,
            fontsize=10, fontweight='bold', va='top')

fig.savefig('figure.pdf', dpi=300, bbox_inches='tight')
```

## matplotlib Subfigures

```python
fig = plt.figure(figsize=(180/25.4, 120/25.4), constrained_layout=True)
subfigs = fig.subfigures(1, 2, width_ratios=[2, 1])

# Left subfigure has 2 stacked panels
axs_left = subfigs[0].subplots(2, 1)
axs_left[0].plot(x, y)
axs_left[1].scatter(x, y, rasterized=True)

# Right subfigure has one panel
ax_right = subfigs[1].subplots(1, 1)
ax_ri

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

pip install matplotlib

git clone https://github.com/GPTomics/bioSkills
git -C bioSkills checkout c8c1fc3a2b13

Works in

1. 2026-10-19
   ✓SAFE B 89 / 100

   first audit · source c8c1fc3a2b13

Questions about Multipanel Figures

• github.com/GPTomics/bioSkills/tree/main/data-visualization/multi — the audited source at commit c8c1fc3a2b13 and every finding aboveObserved · 19 Oct 2026

Spotted something wrong? Report a correction · Methodology