Exact, Paolo. That is what I was looking for and now I understand what was going on: All atoms had coordinates but the hydrogens didn't. To add those, I was embedding the whole thing from scratch, and that was shuffling the coordinates of all atoms.
Thanks, Giammy On Thu, 13 Jan 2022 at 13:25, Paolo Tosco <paolo.tosco.m...@gmail.com> wrote: > Hi Gianmarco, > > you can add hydrogens with coordinates keeping the current heavy atom > coordinates with > > rdkit_mol = rdkit.Chem.AddHs(rdkit_mol, addCoords=True) > > so you may avoid having to call EmbedMolecule, which will compute a whole > set of new coordinates for your molecule. > > I hope I interpreted your needs correctly! > > Cheers, > p. > > > On Thu, Jan 13, 2022 at 2:05 PM Gianmarco Ghiandoni <ghiandon...@gmail.com> > wrote: > >> Hi all, >> >> Quick question: I have noticed that when I read a compound from >> an SDF/Mol file where all coordinates are already defined, the embedding >> process changes the conformation of the RDKit mol object, hence coordinates >> are not preserved: >> >> import rdkit >> cdk2_path = os.path.join(data_path, "cdk2_validation") >> rdkit_mol = rdkit.Chem.MolFromMolFile(os.path.join(cdk2_path, "26Z.sdf")) >> rdkit_mol = rdkit.Chem.AddHs(rdkit_mol) >> rdkit.Chem.AllChem.EmbedMolecule(rdkit_mol) >> >> What is the best practice for reading a molecule with coordinates without >> shuffling them? (e.g. for a ligand from the PDB). >> >> Thanks, >> -- >> *Gianmarco* >> _______________________________________________ >> Rdkit-discuss mailing list >> Rdkit-discuss@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss >> > -- *Gianmarco*
_______________________________________________ Rdkit-discuss mailing list Rdkit-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
