search

forlilab / Meeko

Interfacing RDKit and AutoDock
GNU Lesser General Public License v2.1
192 stars 48 forks source link

The ligand prepared by Meeko for docking using AutoDock-GPU has unusual atoms #158

Closed aprijal99 closed 3 weeks ago

aprijal99 commented 1 month ago

I want to dock a compound from PubChem with ID 11481776 using AutoDock GPU, but encountered an error when generating grid maps using Autogrid4. It said "ERROR: unknown ligand atom type CG0". Then I looked into the compound pdbqt file and found unusual atoms "G"

The pdbqt file of the compound is ok when it was generated by "prepare_ligand" program from ADFR

aprijal99 commented 1 month ago

Screenshot 2024-09-01 070944

rwxayheee commented 3 weeks ago

Hi @aprijal99 These are atom types used for macrocycle sampling (https://autodock-vina.readthedocs.io/en/latest/docking_macrocycle.html) which is supported by Vina 1.2 and AutoDock GPU, but maybe not by ADFR. This function is implemented to sample alternate conformations of (by default, 7 and above)-membered rings in docking calculation. The method was first described in this paper: https://pubs.acs.org/doi/abs/10.1021/ci700036j

If you use ADFR as the docking engine, in ligand preparation you can turn off the macrocycle atom typing using the --rigid_macrocycles option. You can find some examples from the README of this project when you scroll to here: https://github.com/forlilab/Meeko?tab=readme-ov-file#possibly-useful-configurations-of-moleculepreparation

Feel free to reopen the issue if you need any further help