Bio Data Visualization Matplotlib Fundamentals

import matplotlib.pyplot as plt
import matplotlib as mpl

# rcParams for publication compliance
mpl.rcParams.update({
    'pdf.fonttype': 42,                 # TrueType -- searchable PDFs
    'ps.fonttype': 42,                  # TrueType in EPS
    'font.family': 'sans-serif',
    'font.sans-serif': ['Arial', 'Helvetica', 'DejaVu Sans'],
    'font.size': 7,                     # Nature requires 5-7 pt body text
    'axes.labelsize': 7,
    'axes.titlesize': 8,
    'xtick.labelsize': 6,
    'ytick.labelsize': 6,
    'legend.fontsize': 6,
    'figure.dpi': 100,                  # display
    'savefig.dpi': 300,                 # save
    'savefig.bbox': 'tight',
    'savefig.pad_inches': 0.05,
    'axes.linewidth': 0.5,
    'xtick.major.width': 0.5,
    'ytick.major.width': 0.5,
    'lines.linewidth': 1.0,
    'patch.linewidth': 0.5,
})

import matplotlib.pyplot as plt

# Single axes
fig, ax = plt.subplots(figsize=(89/25.4, 70/25.4),       # 89mm x 70mm in inches; Nature single col
                       constrained_layout=True)
ax.scatter(x, y, c='#0072B2', s=10, alpha=0.7, edgecolors='none', rasterized=True)
ax.set_xlabel('PC1 (45%)')
ax.set_ylabel('PC2 (12%)')
ax.spines[['top', 'right']].set_visible(False)
fig.savefig('scatter.pdf')

# Grid of axes
fig, axes = plt.subplots(2, 3, figsize=(180/25.4, 100/25.4),  # 180mm double col
                          constrained_layout=True)
for ax, (label, panel_data) in zip(axes.flat, data.items()):
    ax.plot(panel_data['x'], panel_data['y'])
    ax.set_title(label, fontsize=8)

# Scatter -- always rasterized for >1000 points
ax.scatter(x, y, c=values, cmap='viridis', s=8, alpha=0.6,
           edgecolors='none', rasterized=True)
plt.colorbar(ax.collections[0], ax=ax, label='Expression', shrink=0.8)

# Line
ax.plot(x, y1, color='#0072B2', label='Control', linewidth=1)
ax.plot(x, y2, color='#D55E00', label='Treatment', linewidth=1)
ax.fill_between(x, y_low, y_high, color='#0072B2', alpha=0.2)
ax.legend(frameon=False, fontsize=6)

# Bar
ax.bar(categories, values, color='#0072B2', edgecolor='black', linewidth=0.5)

# Box / violin (prefer seaborn for these -- see distribution-plots)
ax.boxplot([group_a, group_b, group_c], labels=['A', 'B', 'C'],
           patch_artist=True, boxprops=dict(facecolor='#0072B2', alpha=0.7))

# Histogram
ax.hist(values, bins=30, color='#0072B2', edgecolor='white', linewidth=0.5)

# Heatmap (prefer seaborn for clustered; see heatmaps-clustering)
im = ax.imshow(matrix, cmap='RdBu_r', aspect='auto', vmin=-vmax, vmax=vmax)
plt.colorbar(im, ax=ax, label='Z-score')

import seaborn as sns

# seaborn shares the matplotlib Figure/Axes -- pass ax= argument
fig, ax = plt.subplots(figsize=(4, 3), constrained_layout=True)
sns.scatterplot(data=df, x='log_fold_change', y='neg_log_p',
                hue='significance', palette=['#999999', '#0072B2', '#D55E00'],
                s=10, alpha=0.7, ax=ax, rasterized=True)

# seaborn 0.13+ has the `objects` grammar interface (ggplot-like)
import seaborn.objects as so
(so.Plot(df, x='log_fold_change', y='neg_log_p')
   .add(so.Dots(pointsize=2), color='significance')
   .scale(color=['#999999', '#0072B2', '#D55E00']))

# Log scale
ax.set_yscale('log')

# Scientific notation
from matplotlib.ticker import ScalarFormatter
ax.xaxis.set_major_formatter(ScalarFormatter(useMathText=True))

# Date axis
import matplotlib.dates as mdates
ax.xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m'))

# Tick frequency
ax.set_xticks(np.arange(0, 10, 2))
ax.set_xticklabels(['A', 'B', 'C'], rotation=45, ha='right')

# Grid
ax.grid(axis='y', alpha=0.3, linestyle='--', linewidth=0.5)

# CVD-safe categorical
okabe_ito = ['#E69F00', '#56B4E9', '#009E73', '#F0E442',
             '#0072B2', '#D55E00', '#CC79A7', '#000000']

# Perceptually-uniform sequential (Crameri batlow / viridis cividis)
from cmcrameri import cm as cmc
plt.imshow(data, cmap=cmc.batlow)
plt.imshow(data, cmap='viridis')                          # built-in

# Diverging symmetric for LFC / z-score
vmax = np.quantile(np.abs(data), 0.99)
plt.imshow(data, cmap='RdBu_r', vmin=-vmax, vmax=vmax)   # symmetric

# PDF for vector text + raster scatter (best of both)
fig.savefig('figure.pdf', dpi=300, bbox_inches='tight')

# PNG for raster (web, presentations)
fig.savefig('figure.png', dpi=300, bbox_inches='tight')

# TIFF for some journals
fig.savefig('figure.tiff', dpi=300, pil_kwargs={'compression': 'tiff_lzw'})

# SVG for editable vector
fig.savefig('figure.svg', bbox_inches='tight')