plt_histo.py

import numpy as np
import matplotlib.pyplot as plt


histo = np.loadtxt('histo.xvg', comments=['#', '@'])

x_values = histo[:, 0]

plt.figure(figsize=(10, 4))

for hist in histo[:, 1:].T:
    hist = [float('nan') if x==0 else x for x in hist] # Don't plot zero values
    plt.plot(x_values, hist, '.-')

plt.xlabel('COM separation, nm')
plt.ylabel('Histogram counts, a.u.')
plt.savefig('histo.pdf')

Context: PMFs obtained via umbrella sampling in GROMACS rely on inverting histograms with WHAM. GROMACS is great for preparing, running, and analyzing these simulations. However, the raw histogram data are thrown to xvg files, which requires not only xmgrace to be installed and working properly, but some magic to actually plot all columns at once. This snippet avoids that.