search

forlilab / Meeko

Interfacing RDKit and AutoDock
GNU Lesser General Public License v2.1
169 stars 41 forks source link

LinkerRDKitChorizo cannot process PRO residue #136

Open HaoLan-compchem opened 1 week ago

HaoLan-compchem commented 1 week ago

I tried to use mk_prepare_receptor.py for preparing protein grid before ligand docking.

Here is an error given by meeko:

"C:PRO:357": "C:PRO:357" } Traceback (most recent call last): File "/data/home/hlan/.conda/envs/bd_env/bin/mk_prepare_receptor.py", line 332, in chorizo = LinkedRDKitChorizo(pdb_string, mutate_res_dict=mutate_res_dict, termini=termini, deleted_residues=del_res, allow_bad_res=args.allow_bad_res, skip_auto_disulfide=args.skip_auto_disulfide) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/data/home/hlan/.conda/envs/bd_env/lib/python3.11/site-packages/meeko/linked_rdkit_chorizo.py", line 260, in init raise RuntimeError(msg) RuntimeError: The following residues could not be processed: Resname: PRO: C:231 C:357

Here is one PRO residue data (with protons) in my PDB file:

ATOM 11226 N PRO C 357 -43.712 -25.058 11.337 1.00 0.00 A N ATOM 11227 CA PRO C 357 -43.210 -23.926 12.134 1.00 0.00 A C ATOM 11228 C PRO C 357 -42.343 -22.913 11.363 1.00 0.00 A C ATOM 11229 O PRO C 357 -41.303 -23.272 10.801 1.00 0.00 A O ATOM 11230 CB PRO C 357 -42.407 -24.544 13.277 1.00 0.00 A C ATOM 11231 CG PRO C 357 -43.095 -25.897 13.491 1.00 0.00 A C ATOM 11232 CD PRO C 357 -43.478 -26.300 12.065 1.00 0.00 A C ATOM 11233 HA PRO C 357 -44.062 -23.384 12.570 1.00 0.00 A H ATOM 11234 1HB PRO C 357 -42.433 -23.933 14.179 1.00 0.00 A H ATOM 11235 2HB PRO C 357 -42.842 -25.515 13.560 1.00 0.00 A H ATOM 11236 1HG PRO C 357 -43.989 -25.757 14.087 1.00 0.00 A H ATOM 11237 2HG PRO C 357 -43.412 -26.319 12.525 1.00 0.00 A H ATOM 11238 1HD PRO C 357 -44.370 -26.922 12.108 1.00 0.00 A H ATOM 11239 2HD PRO C 357 -43.653 -25.396 11.463 1.00 0.00 A H

My conda environment version of meeko and rdkit:

meeko 0.6.0a3 pypi_0 pypi rdkit 2023.9.5 pypi_0 pypi

I have also tried to downgrade meeko to 0.5.0 which is even worse (process much less residues excluding HIS...)

So I am wondering if it could an issue of the version of rdkit Chem.MolFromPDBBlock module which is called in the LinkerRDKitChorizo.py script? Is there a way to process the proline residue with protons in above format?

Thank you very much.

diogomart commented 3 days ago

Hi,

the hydrogens are messed up, v0.6.0a3 should work if the hydrogens are fixed or removed pro357