visualization

# Scientific Visualization

Publication-quality figures with Python. Use venv: `source /Users/zhangmingda/clawd/.venv/bin/activate`

## Style Defaults (journal-ready)

```python
import matplotlib.pyplot as plt
import matplotlib as mpl
import seaborn as sns
import numpy as np

# Publication style
plt.rcParams.update({
    'font.size': 12,
    'font.family': 'sans-serif',
    'axes.linewidth': 1.2,
    'axes.spines.top': False,
    'axes.spines.right': False,
    'figure.dpi': 150,
    'savefig.dpi': 300,
    'savefig.bbox': 'tight',
    'savefig.transparent': True,
})
sns.set_palette("colorblind")  # accessible colors
```

## Common Plot Types

### Distribution
```python
fig, ax = plt.subplots(figsize=(6, 4))
sns.histplot(data=df, x='value', hue='group', kde=True, ax=ax)
ax.set_xlabel('Value')
ax.set_ylabel('Count')
plt.savefig('dist.png', dpi=300)
```

### Comparison (box + strip)
```python
fig, ax = plt.subplots(figsize=(6, 4))
sns.boxplot(data=df, x='group', y='value', ax=ax, width=0.5)
sns.stripplot(data=df, x='group', y='value', ax=ax, color='black', alpha=0.3, size=3)
ax.set_ylabel('Measurement (units)')
plt.savefig('comparison.png', dpi=300)
```

### Scatter + Regression
```python
fig, ax = plt.subplots(figsize=(6, 5))
sns.regplot(data=df, x='x', y='y', ax=ax, scatter_kws={'alpha': 0.5})
r, p = stats.pearsonr(df['x'], df['y'])
ax.annotate(f'r = {r:.3f}, p = {p:.3g}', xy=(0.05, 0.95), xycoords='axes fraction', fontsize=10)
plt.savefig('scatter.png', dpi=300)
```

### Heatmap (correlation / expression)
```python
fig, ax = plt.subplots(figsize=(8, 6))
sns.heatmap(corr_matrix, annot=True, fmt='.2f', cmap='RdBu_r', center=0,
            square=True, linewidths=0.5, ax=ax)
plt.savefig('heatmap.png', dpi=300)
```

### Multi-panel Figure
```python
fig, axes = plt.subplots(1, 3, figsize=(15, 4))
# Panel A
axes[0].plot(x, y)
axes[0].set_title('A', loc='left', fontweight='bold')
# Panel B
axes[1].bar(categories, values)
axes[1].set_title('B', loc='left', fontweight='bold')
# Panel C
axes[2].scatter(x2, y2)
axes[2].set_title('C', loc='left', fontweight='bold')
plt.tight_layout()
plt.savefig('figure1.png', dpi=300)
```

### Volcano Plot (genomics)
```python
fig, ax = plt.subplots(figsize=(7, 5))
colors = np.where((df['padj'] < 0.05) & (abs(df['log2FC']) > 1), 'red',
         np.where(df['padj'] < 0.05, 'blue', 'grey'))
ax.scatter(df['log2FC'], -np.log10(df['padj']), c=colors, alpha=0.5, s=10)
ax.axhline(-np.log10(0.05), ls='--', color='grey', lw=0.8)
ax.axvline(-1, ls='--', color='grey', lw=0.8)
ax.axvline(1, ls='--', color='grey', lw=0.8)
ax.set_xlabel('log₂ Fold Change')
ax.set_ylabel('-log₁₀ adjusted p-value')
plt.savefig('volcano.png', dpi=300)
```

### Network Graph
```python
import networkx as nx
G = nx.from_pandas_edgelist(df, 'source', 'target', 'weight')
pos = nx.spring_layout(G, seed=42)
fig, ax = plt.subplots(figsize=(8, 8))
nx.draw_networkx(G, pos, ax=ax, node_size=300, font_size=8, edge_color='grey', alpha=0.7)
plt.savefig('network.png', dpi=300)
```

### Interactive (Plotly)
```python
import plotly.express as px
fig = px.scatter(df, x='x', y='y', color='group', hover_data=['label'],
                 title='Interactive Scatter')
fig.write_html('interactive.html')
fig.write_image('scatter.png', scale=2)  # needs kaleido
```

## Journal Requirements

| Journal | Width (single col) | Width (double col) | Format | Font min |
|---------|-------------------|-------------------|--------|----------|
| Nature | 89mm | 183mm | PDF/EPS/TIFF | 5pt |
| Science | 85mm | 174mm | PDF/EPS | 6pt |
| PNAS | 87mm | 178mm | PDF/EPS/TIFF | 6pt |
| IEEE | 3.5in | 7.16in | PDF/EPS | 8pt |
| Elsevier | 90mm | 190mm | PDF/EPS/TIFF | 6pt |

```python
# Nature single-column figure
fig, ax = plt.subplots(figsize=(3.5, 2.6))  # 89mm ≈ 3.5in
```

## Accessibility
- Use colorblind-safe palettes: `sns.set_palette("colorblind")`
- Add patterns/markers in addition to color
- Ensure sufficient contrast
- Use descriptive axis labels with units
- Include alt text in figure captions

## Tips
- Save as both PNG (for preview) and PDF/SVG (for publication)
- Always label axes with units
- Use consistent color coding across related figures
- Avoid 3D plots unless data is truly 3D
- Minimize chart junk (unnecessary gridlines, borders, decorations)