Tutorial: High-Quality Figure Generation¶

This tutorial demonstrates how to generate high-quality figures for your research.

Figure Quality Standards¶

ASCICat generates figures meeting scientific standards:

Specification	Value
Resolution	600 DPI
Formats	PNG, PDF, SVG
Color scheme	Colorblind-friendly
Font	Sans-serif (Arial-like)

The Four-Panel Figure¶

The signature ASCICat figure consists of four panels:

from ascicat import ASCICalculator
from ascicat.visualizer import Visualizer
from pathlib import Path

# Setup
output_dir = Path('figures/output')
output_dir.mkdir(parents=True, exist_ok=True)

# Calculate ASCI
calc = ASCICalculator(reaction='HER')
calc.load_data('data/HER_clean.csv')
results = calc.calculate_asci(w_a=0.33, w_s=0.33, w_c=0.34)

# Initialize visualizer
viz = Visualizer(results, calc.config, auto_sample=True)

# Generate all panels
viz.generate_figures(
    output_dir=str(output_dir),
    dpi=600,
    formats=['png', 'pdf']
)

Panel A: 3D ASCI Component Space¶

fig = viz.plot_3d_component_space(
    highlight_top_n=10,
    figsize=(8, 7),
    elev=25,           # Elevation angle
    azim=45,           # Azimuth angle
    marker_size=50,
    alpha=0.6
)
fig.savefig(output_dir / 'panel_a.png', dpi=600, bbox_inches='tight')
fig.savefig(output_dir / 'panel_a.pdf', bbox_inches='tight')

Panel B: Rank vs. Adsorption Energy¶

fig = viz.plot_rank_vs_adsorption(
    show_optimal=True,      # Mark optimal energy
    show_window=True,       # Show activity window
    n_highlight=10,         # Label top catalysts
    figsize=(8, 6)
)
fig.savefig(output_dir / 'panel_b.png', dpi=600, bbox_inches='tight')

Panel C: Volcano Optimization Landscape¶

fig = viz.plot_volcano_optimization(
    n_contours=20,
    cmap='plasma',
    figsize=(8, 6),
    show_annotations=True
)
fig.savefig(output_dir / 'panel_c.png', dpi=600, bbox_inches='tight')

Panel D: Top Performers Breakdown¶

fig = viz.plot_top_performers(
    n_show=15,
    figsize=(10, 6),
    show_values=True
)
fig.savefig(output_dir / 'panel_d.png', dpi=600, bbox_inches='tight')

Customizing Figures¶

Color Schemes¶

# Colorblind-friendly palette (default)
viz = Visualizer(results, calc.config, color_palette='colorblind')

# Alternative palettes
viz = Visualizer(results, calc.config, color_palette='deep')
viz = Visualizer(results, calc.config, color_palette='muted')

Custom Colors¶

import matplotlib.pyplot as plt

fig = viz.plot_top_performers(n_show=15)
ax = fig.axes[0]

# Customize colors for score components
# Activity = green, Stability = blue, Cost = orange
for patch in ax.patches[:15]:
    patch.set_facecolor('#2ca02c')  # Activity bars

plt.tight_layout()
fig.savefig('custom_colors.png', dpi=600)

Font Sizes¶

import matplotlib.pyplot as plt

# Set global font sizes
plt.rcParams.update({
    'font.size': 12,
    'axes.titlesize': 14,
    'axes.labelsize': 12,
    'xtick.labelsize': 10,
    'ytick.labelsize': 10,
    'legend.fontsize': 10
})

fig = viz.plot_asci_distribution()
fig.savefig('custom_fonts.png', dpi=600)

Supplementary Figures¶

ASCI Score Distribution¶

fig = viz.plot_asci_distribution(
    bins=50,
    show_statistics=True,
    figsize=(8, 5)
)

# Add annotations
ax = fig.axes[0]
ax.axvline(results['ASCI'].median(), color='red',
           linestyle='--', label='Median')
ax.legend()

fig.savefig(output_dir / 'supp_distribution.png', dpi=600)

Score Correlation Matrix¶

fig = viz.plot_score_correlations(figsize=(8, 7))
fig.savefig(output_dir / 'supp_correlations.png', dpi=600)

Pareto Frontier (2D)¶

# Activity vs Cost trade-off
fig = viz.plot_pareto_2d(
    x_score='activity',
    y_score='cost',
    highlight_top_n=20,
    figsize=(8, 6)
)
fig.savefig(output_dir / 'supp_pareto_activity_cost.png', dpi=600)

# Activity vs Stability trade-off
fig = viz.plot_pareto_2d(
    x_score='activity',
    y_score='stability',
    highlight_top_n=20
)
fig.savefig(output_dir / 'supp_pareto_activity_stability.png', dpi=600)

Radar Chart for Top Catalysts¶

# Compare score profiles
top_symbols = results.head(5)['symbol'].tolist()

fig = viz.plot_radar_chart(
    catalyst_symbols=top_symbols,
    figsize=(8, 8)
)
fig.savefig(output_dir / 'supp_radar.png', dpi=600)

Interactive Figures¶

For presentations or supplementary materials:

# Interactive 3D plot (HTML)
viz.create_interactive_3d(
    output_path=str(output_dir / 'interactive_3d.html'),
    title='HER Catalyst Screening Results',
    highlight_top_n=20
)

Figure Legends¶

Suggested Panel Descriptions¶

**Figure 1.** ASCI-based HER catalyst screening results.
**(A)** Three-dimensional visualization of catalyst scores in the
Activity-Stability-Cost space. Points are colored by ASCI score
(colorbar). Stars indicate the top 10 ranked catalysts.
**(B)** ASCI rank versus hydrogen adsorption energy. The red dashed
line marks the Sabatier optimal energy (−0.27 eV). The green shaded
region indicates the activity scoring window (±σ_a).
**(C)** Volcano optimization landscape showing ASCI contours over the
energy-cost parameter space. The optimal region combines near-thermoneutral
binding with low material cost.
**(D)** Score decomposition for the top 15 catalysts. Bar segments show
contributions from activity (green), stability (blue), and cost (orange)
components. Black markers indicate overall ASCI scores.

Export Settings¶

High Resolution Output¶

# Single column width: 89 mm
# Double column width: 183 mm

fig = viz.plot_top_performers(n_show=10)
fig.set_size_inches(3.5, 3)  # Single column
fig.savefig('single_column.pdf', dpi=600, bbox_inches='tight')

fig = viz.plot_3d_component_space()
fig.set_size_inches(7.2, 5)  # Double column
fig.savefig('double_column.pdf', dpi=600, bbox_inches='tight')

Standard Sizes¶

# Single column: 3.25 inches
# Double column: 7 inches

fig = viz.plot_asci_distribution()
fig.set_size_inches(3.25, 2.5)
fig.savefig('standard_single.pdf', dpi=600, bbox_inches='tight')

Complete Figure Script¶

#!/usr/bin/env python3
"""Generate all figures for HER screening."""

from ascicat import ASCICalculator
from ascicat.visualizer import Visualizer
from pathlib import Path
import matplotlib.pyplot as plt

# Configure matplotlib
plt.rcParams.update({
    'font.family': 'sans-serif',
    'font.size': 10,
    'axes.linewidth': 1.0,
    'xtick.major.width': 1.0,
    'ytick.major.width': 1.0
})

def main():
    output = Path('figures/output')
    output.mkdir(parents=True, exist_ok=True)

    # Calculate
    calc = ASCICalculator(reaction='HER')
    calc.load_data('data/HER_clean.csv')
    results = calc.calculate_asci()

    # Visualize
    viz = Visualizer(results, calc.config)

    # Main figure panels
    viz.generate_figures(
        output_dir=str(output),
        dpi=600,
        formats=['png', 'pdf']
    )

    # Supplementary
    viz.plot_asci_distribution().savefig(
        output / 'supp_distribution.pdf', dpi=600)
    viz.plot_score_correlations().savefig(
        output / 'supp_correlations.pdf', dpi=600)

    # Interactive
    viz.create_interactive_3d(str(output / 'interactive.html'))

    print(f"Figures saved to: {output}/")

if __name__ == '__main__':
    main()