Using compound names instead of pose number as X axis of interaction barcode.

[Le-Phung-Hien]

Hi,

I am running a script to get interaction barcode of different compounds (in a single sdf file) to a protein (in a pdb file).

The script run well, but I would like to know if there is a way to use compound names instead of pose number as X axis of interaction barcode? Also, is it possible to use 1-letter code instead of 3-letter code for the residues?

Thank you very much.

Hien

import MDAnalysis as mda
import prolif as plf
import sys
import argparse
import os
from IPython.display import display
from matplotlib import pyplot as plt
from rdkit import Chem

# Create an argument parser
parser = argparse.ArgumentParser(description='Protein-ligand interaction barcoding')

# Add arguments
parser.add_argument('-p', '--protein', type=str, help='Path to the protein file')
parser.add_argument('-l', '--ligands', type=str, help='Path to the ligand file')
parser.add_argument('-o', '--output', type=str, help='Output image name')
group = parser.add_mutually_exclusive_group(required=True)
group.add_argument("-md", "--mdanalysis", action="store_true", help='Using mdanalysis to parse protein')
group.add_argument("-rd", "--rdkit", action="store_true", help='Using rdkit to parse protein')
# Parse the command-line arguments
args = parser.parse_args()

protein = args.protein
ligands = args.ligands
output = args.output
# Read the protein file
if args.mdanalysis:
    u = mda.Universe(protein)
    protein_mol = plf.Molecule.from_mda(u)
elif args.rdkit: 
    rdkit_prot = Chem.MolFromPDBFile(protein, removeHs=False)
    protein_mol = plf.Molecule(rdkit_prot)
# Load ligands
poses_path = str(os.path.abspath(ligands))
if ligands.endswith('.sdf'):
    pose_iterable = plf.sdf_supplier(poses_path)
if ligands.endswith('.mol2'):
    pose_iterable = plf.mol2_supplier(poses_path)

# Use default interactions
fp = plf.Fingerprint()

# Run on your poses
fp.run_from_iterable(pose_iterable, protein_mol)

# show dataframe
df = fp.to_dataframe(index_col="Compound")
display(df)

# show barcode
fp.plot_barcode(xlabel="Pose")
plt.savefig(output, dpi=600)
plt.show()

# show first compound's interactions
fp.plot_lignetwork(pose_iterable[0])

[cbouy]

Hi Hien,

fp.plot_barcode(xlabel="Pose") returns a matplotlib Axes object from which you can manipulate the xticks and xticklabels attributes to do what you want (see their docs here).

As for switching the residue names to be 1-letter code, you can extract the current labels with the same attributes as above (just y instead of x), and if you extract the first 3 characters you can then use MDAnalysis to convert the 3-letter code to 1-letter with MDAnalysis.lib.util.inverse_aa_codes or your own conversion table.