How to fix the label clashing with itself

[vishal9506]

Hi,

I have been trying to use chiLife to generate distance distribution of my own label on a protein. Unfortunately, when I have a look at the individual label conformation, I could see some conformations which is just not possible (label clashing with itself). Could someone please help me fix this issue.

Example:

The script that I use:

import os
import numpy as np
import matplotlib.pyplot as plt
import chilife as chi
import MDAnalysis as mda

np.random.seed(100)
plt.style.use('chiLife')
plt.rcParams.update({'font.size':15})

os.chdir('C:\Users\uqvbayya\Desktop\Chilife\VBGd1_cysCNP\G147')

MBP = mda.Universe('C:/Users/uqvbayya/Desktop/Chilife/Protein/ERp29.pdb')

SL1 = chi.RotamerEnsemble('label', site=147, chain='A', protein=MBP, sample=10000, eval_clash=True)
SL2 = chi.RotamerEnsemble('label', site=403, chain='B', protein=MBP, sample=10000, eval_clash=True)

chi.save('ERp29_labelled.pdb', SL1, SL2, MBP)

traj, dE = chi.repack(MBP, SL1, SL2,
repetitions=10000, temp=298,
off_rotamer=True,
repack_radius=10)

SL1_repack = chi.SpinLabel.from_trajectory(traj, 147, burn_in=5000, spin_atoms=SL1.spin_atoms)
SL2_repack = chi.SpinLabel.from_trajectory(traj, 403, burn_in=5000, spin_atoms=SL2.spin_atoms)

chi.save('ERp29_labelled_repacked.pdb', SL1_repack, SL2_repack, MBP)

fig, ax = plt.subplots()
ax.plot(np.cumsum(dE))
ax.set_xlabel('step')
ax.set_ylabel(r'$\rm\Delta$ clash energy (kcal/mol)')
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)
plt.show()

fig.savefig('Trajectory.png', bbox_inches='tight', dpi=1200)

r = np.linspace(0, 100, 256)
P_repack = chi.distance_distribution(SL1_repack, SL2_repack, r)

fig, ax = plt.subplots(figsize=(6, 3))
ax.plot(r, P_repack)
ax.set_yticks([])
ax.set_xlabel('Distance ($\AA$)')

for spine in ['left', 'right', 'top']:
ax.spines[spine].set_visible(False)
plt.show()

fig.savefig('Distance_distributions.png', dpi=1200, bbox_inches='tight')

distancedistribution = np.column_stack((r, P_repack))
np.savetxt('distancedistribution.txt', distancedistribution, header="Distance[nm] Probability")

Kind regards,
Vishal

[mtessmer]

Hi Vishal, can you provide me (by email: mhtessmer@gmail.com) with the PDB file of your custom label and the code you used to create the rotamer library or the rotlib file itself? In theory, chilife shouldn't sample any dihedrals that are part of rings, but this may depend on how the initial rotamer library was constructed.