Publication-Ready Figures

SkillDev tools

Generate publication-quality figures following journal standards for Nature, Cell, and other high-impact journals. Use when user wants to create figures for papers, presentations, or reports with professional formatting. Triggers: "publication figure", "paper figure", "Nature style", "journal figure", "export for publication", "figure panel", "high quality plot".

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Publication-Ready Figures skill

What this skill tells your AI

The instructions your AI receives, as published by cafferychen777/chatspatial in .agents/skills/publication-ready/SKILL.md and read by ahel’s review.

Overview

This skill creates publication-quality visualizations following journal standards.

Key principles:

  1. Clarity over decoration: Every element serves a purpose
  2. Consistent styling: Uniform fonts, colors, and sizing
  3. Reproducibility: Settings documented for revision cycles

Journal Standards Quick Reference

Figure Dimensions

JournalSingle Column1.5 ColumnDouble Column
Nature89 mm120 mm183 mm
Cell85 mm114 mm174 mm
Science55 mm117 mm183 mm

Height: Typically 1:1 to 1:1.5 aspect ratio; max ~230mm

Resolution Requirements

FormatResolutionUse Case
Print300-600 DPIFinal submission
Vector (PDF/SVG)N/APreferred for line art
Web/Review150 DPIInitial submission

Font Standards

Recommended fonts:
- Sans-serif: Arial, Helvetica (most common)
- Serif: Times New Roman (less common for figures)

Size guidelines:
- Axis labels: 7-8 pt
- Tick labels: 6-7 pt
- Panel labels (a, b, c): 8-10 pt, bold
- Annotations: 6-8 pt

Color Guidelines

Colorblind-Friendly Palettes

Recommended categorical palette:

Blue:    #0072B2
Orange:  #E69F00
Green:   #009E73
Yellow:  #F0E442
Sky:     #56B4E9
Red:     #D55E00
Pink:    #CC79A7
Gray:    #999999

Sequential palettes:

  • Expression: viridis, magma, inferno
  • Diverging: RdBu (centered at 0), coolwarm

Color Usage Rules

  1. Maximum 7-8 colors for categorical data
  2. Consistent color mapping across all panels
  3. Avoid red-green combinations (colorblind unfriendly)
  4. White background for most journals

Multi-Panel Figure Layout

Panel Organization

Standard layouts:

2-panel: [a][b]

3-panel: [a][b]    or    [a][b][c]
         [c]

4-panel: [a][b]
         [c][d]

6-panel: [a][b][c]
         [d][e][f]

Panel Labels

  • Position: Top-left corner of each panel
  • Style: Lowercase bold (a, b, c) or uppercase (A, B, C)
  • Font: Same as figure, 8-10 pt bold

Visualization Type Guidelines

Spatial Plots

Essential elements:
- Scale bar (required for spatial)
- Clear spot/cell boundaries
- Consistent color scale across samples

Avoid:
- Overcrowded labels
- Low contrast colors
- Missing coordinate axes

Heatmaps

Essential elements:
- Dendrogram if clustered
- Row/column labels (or clear groupings)
- Color scale bar with range

Settings:
- Center diverging scales at 0 or mean
- Use appropriate normalization (z-score for expression)

UMAP/Embedding Plots

Essential elements:
- Legend (positioned outside if crowded)
- Point size appropriate for density
- Consistent coloring with spatial plots

Avoid:
- Axis labels (UMAP axes are arbitrary)
- Grid lines
- Cluttered legends

Violin/Box Plots

Essential elements:
- Y-axis label with units
- Statistical comparisons if relevant
- Consistent colors with other panels

Good practices:
- Order groups logically
- Add individual points for small n
- Include median/quartile indicators

Export Workflow

Step 1: Configure Visualization

Use visualize_data with:

params:
  output_format: "pdf"  # Vector format preferred
  dpi: 300              # For raster elements
  figsize: [3.5, 3.5]   # In inches (89mm ≈ 3.5in)

Step 2: Export Options

FormatWhen to UseFile Size
PDFVector graphics, line artSmall
SVGWeb, editable graphicsSmall
PNGRaster, screenshotsMedium
TIFFPrint submission (some journals)Large
EPSLegacy systemsSmall

Step 3: Post-Processing (if needed)

For complex multi-panel figures:

  1. Export individual panels as PDF/SVG
  2. Assemble in vector software (Illustrator, Inkscape)
  3. Add panel labels and annotations
  4. Export final composite

Common Figure Types for Spatial Analysis

Figure 1: Data Overview

Typical panels:
a) Spatial scatter - clusters/domains
b) UMAP embedding
c) Violin plot - key markers
d) Spatial expression - signature genes

Figure 2: Cell Composition

Typical panels:
a) Deconvolution - spatial proportions
b) Reference UMAP with cell types
c) Individual cell type maps
d) Proportion bar plots by region

Figure 3: Cell Communication

Typical panels:
a) Circle plot - communication network
b) Dotplot - top interactions
c) Spatial L-R pairs
d) Pathway analysis

Figure 4: Dynamics

Typical panels:
a) Velocity stream plot
b) Pseudotime trajectory
c) Gene trends along trajectory
d) Fate probability map

Quality Checklist

Before submission:

  • All text is legible at print size
  • Colors are consistent across panels
  • Scale bars present on spatial plots
  • Legends are complete and clear
  • File format matches journal requirements
  • Resolution meets minimum standards
  • Panel labels are properly positioned
  • No overlapping elements

Troubleshooting

IssueSolution
Text too smallIncrease figure size or font size
Colors muddy in printTest grayscale conversion
File too largeUse vector format, reduce DPI for raster
Legend overlaps dataMove legend outside plot area
Inconsistent stylingCreate style template, apply to all

Signals

GitHub stars
44
Forks
13
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
publication-ready
Source
github.com/cafferychen777/chatspatial