brycewang-stanford/scientific-illustration-guide

Scientific Illustration Guide

A skill for creating clear, professional scientific illustrations including graphical abstracts, schematic diagrams, workflow visualizations, and architecture diagrams. Covers both programmatic and design tool approaches.

Graphical Abstract Design

Composition Principles

Layout Guidelines for Graphical Abstracts:
  - Dimensions: typically 500x300px to 1200x800px (check journal spec)
  - Flow direction: left-to-right or top-to-bottom
  - Maximum 5-7 visual elements
  - Use arrows to show process flow
  - Include 1-2 key data points or results
  - Minimal text (30-50 words maximum)
  - Consistent color scheme (3-4 colors)

Structure:
  [Input/Problem] --> [Method/Process] --> [Output/Finding]

Programmatic Diagrams with Python

import matplotlib.pyplot as plt
import matplotlib.patches as patches
from matplotlib.patches import FancyBboxPatch, FancyArrowPatch

def create_workflow_diagram(steps: list[dict], output: str = 'workflow.pdf'):
    """
    Create a horizontal workflow diagram.

    Args:
        steps: List of dicts with 'label', 'color', and optional 'sublabel'
        output: Output file path
    """
    fig, ax = plt.subplots(figsize=(12, 3))
    n = len(steps)
    box_width = 0.12
    gap = (1 - n * box_width) / (n + 1)

    for i, step in enumerate(steps):
        x = gap + i * (box_width + gap)
        y = 0.3

        # Draw box
        box = FancyBboxPatch(
            (x, y), box_width, 0.4,
            boxstyle="round,pad=0.01",
            facecolor=step.get('color', '#3B82F6'),
            edgecolor='#1E293B',
            linewidth=1.5,
            alpha=0.9
        )
        ax.add_patch(box)

        # Add label
        ax.text(x + box_width/2, y + 0.2, step['label'],
                ha='center', va='center', fontsize=9,
                fontweight='bold', color='white')

        if 'sublabel' in step:
            ax.text(x + box_width/2, y - 0.08, step['sublabel'],
                    ha='center', va='center', fontsize=7, color='#475569')

        # Draw arrow to next step
        if i < n - 1:
            ax.annotate('', xy=(x + box_width + gap*0.2, 0.5),
                        xytext=(x + box_width + gap*0.8, 0.5),
                        arrowprops=dict(arrowstyle='->', color='#64748B',
                                       lw=1.5))

    ax.set_xlim(0, 1)
    ax.set_ylim(0, 1)
    ax.axis('off')
    fig.savefig(output, dpi=300, bbox_inches='tight',
                facecolor='white', edgecolor='none')
    plt.close()
    return output

# Example: research pipeline
steps = [
    {'label': 'Data\nCollection', 'color': '#3B82F6', 'sublabel': 'N=1,200'},
    {'label': 'Preprocessing', 'color': '#6366F1', 'sublabel': 'QC + Filtering'},
    {'label': 'Analysis', 'color': '#8B5CF6', 'sublabel': 'ML Pipeline'},
    {'label': 'Validation', 'color': '#A855F7', 'sublabel': 'Cross-validation'},
    {'label': 'Results', 'color': '#EC4899', 'sublabel': 'AUC = 0.92'}
]
create_workflow_diagram(steps, 'research_pipeline.pdf')

Schematic Diagrams

System Architecture Diagrams

import matplotlib.pyplot as plt
from matplotlib.patches import FancyBboxPatch, FancyArrowPatch

def create_architecture_diagram(components: list[dict],
                                  connections: list[tuple],
                                  output: str = 'architecture.pdf'):
    """
    Create a system architecture diagram.

    Args:
        components: List of {'name', 'x', 'y', 'width', 'height', 'color', 'type'}
        connections: List of (source_name, target_name, label) tuples
    """
    fig, ax = plt.subplots(figsize=(10, 7))

    # Draw components
    comp_positions = {}
    for comp in components:
        x, y = comp['x'], comp['y']
        w, h = comp.get('width', 1.5), comp.get('height', 0.8)
        color = comp.get('color', '#3B82F6')

        if comp.get('type') == 'database':
            # Cylinder shape for databases
            ellipse_h = 0.15
            ax.add_patch(patches.Rectangle((x, y), w, h-ellipse_h,
                         facecolor=color, edgecolor='#1E293B', linewidth=1.2))
            ax.add_patch(patches.Ellipse((x+w/2, y+h-ellipse_h), w, ellipse_h*2,
                         facecolor=color, edgecolor='#1E293B', linewidth=1.2))
            ax.add_patch(patches.Ellipse((x+w/2, y), w, ellipse_h*2,
                         facecolor=color, edgecolor='#1E293B', linewidth=1.2))
        else:
            box = FancyBboxPatch((x, y), w, h,
                                  boxstyle="round,pad=0.05",
                                  facecolor=color, edgecolor='#1E293B',
                                  linewidth=1.2, alpha=0.9)
            ax.add_patch(box)

        ax.text(x + w/2, y + h/2, comp['name'],
                ha='center', va='center', fontsize=10,
                fontweight='bold', color='white')

        comp_positions[comp['name']] = (x + w/2, y + h/2, w, h)

    # Draw connections
    for src, tgt, label in connections:
        sx, sy, sw, sh = comp_positions[src]
        tx, ty, tw, th = comp_positions[tgt]

        ax.annotate('', xy=(tx, ty + th/2 if sy > ty else ty - th/2),
                    xytext=(sx, sy - sh/2 if sy > ty else sy + sh/2),
                    arrowprops=dict(arrowstyle='->', color='#64748B',
                                   lw=1.5, connectionstyle='arc3,rad=0.1'))

        mid_x = (sx + tx) / 2
        mid_y = (sy + ty) / 2
        if label:
            ax.text(mid_x + 0.1, mid_y, label, fontsize=7, color='#475569')

    ax.set_xlim(-0.5, 10)
    ax.set_ylim(-0.5, 8)
    ax.set_aspect('equal')
    ax.axis('off')
    fig.savefig(output, dpi=300, bbox_inches='tight',
                facecolor='white')
    plt.close()

Vector Graphics with Drawio

For complex diagrams, use draw.io (diagrams.net) which exports to SVG, PDF, and PNG:

<!-- Example draw.io XML structure -->
<mxGraphModel>
  <root>
    <mxCell id="0"/>
    <mxCell id="1" parent="0"/>
    <mxCell id="2" value="Data Source" style="rounded=1;fillColor=#3B82F6;
            fontColor=#FFFFFF;strokeColor=#1E40AF;" vertex="1" parent="1">
      <mxGeometry x="80" y="80" width="120" height="60" as="geometry"/>
    </mxCell>
  </root>
</mxGraphModel>

Recommended Tools by Diagram Type

| Diagram Type | Best Tool | Output Format | Learning Curve | |-------------|-----------|---------------|----------------| | Flowcharts | draw.io / Mermaid | SVG, PDF | Low | | Molecular structures | ChemDraw / RDKit | SVG, PNG | Medium | | Biological pathways | BioRender / KEGG | PNG, SVG | Low | | Network graphs | Cytoscape / NetworkX | SVG, PDF | Medium | | 3D protein structures | PyMOL / ChimeraX | PNG, TIFF | High | | Circuit diagrams | KiCad / CircuiTikZ | PDF, SVG | Medium | | Math diagrams | TikZ/PGF | PDF | High | | UML diagrams | PlantUML / Mermaid | SVG, PNG | Low |

Color and Accessibility

Use colorblind-safe palettes consistently. The key rules:

1. Never use red and green together to encode different categories
2. Use both color and shape/pattern to distinguish elements
3. Ensure sufficient contrast (WCAG AA: 4.5:1 ratio for text)
4. Test your diagram in grayscale to verify it remains interpretable
5. Label elements directly rather than relying solely on a color legend

Export and Journal Compliance

• Export vector formats (SVG, PDF, EPS) for line art and diagrams
• Minimum 300 DPI for raster elements within diagrams
• Embed fonts or convert text to paths in final SVG/PDF
• Check journal-specific requirements for graphical abstract dimensions and file size
• Name files following journal convention (e.g., "Figure_1.pdf", "graphical_abstract.tiff")