forlilab / Meeko

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

Water bond angle not correct #100

Open althea-hansel opened 2 months ago

althea-hansel commented 2 months ago

I am giving meeko a receptor to prepare with water oxygen positions only, allowing it to add hydrogens to the waters. The resulting bond angle for the waters in the PDBQT is 109.5°, not 104.5°

diogomart commented 2 months ago

RDKit adds the hydrogens, and it uses a minimal set of rules to specify the geometry. I think that we need to minimize the hydrogens with a force field like OpenFF or Espaloma.

That's probably not a big deal for the calculation, in view of the quality of the partial charges, lack of polarization, and modern water models using angles closer to 109 than to 105 deg (OPC3 and TIP3P-FB) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4991989/table/t2/?report=objectonly

althea-hansel commented 2 months ago

So, leave it as is?

diogomart commented 2 months ago

yes