On FF parameterization of molecules with non-zero valence

User discussions
1

Hello all,

I’m working on a flavonoid that happens to have a net non-zero valence (one of the oxygen has 3 bonds, so +1 positive charge):

SMILES: C[C@H]1[C@@H]([C@H]([C@H]([C@@H](O1)OC[C@@H]2[C@H]([C@@H]([C@H]([C@@H](O2)OC3=CC4=C(C=C(C=C4[O+]=C3C5=CC(=C(C=C5)O)O)O)O)O)O)O)O)O)O

Since FF parameterization works only if the net valence of the molecule is zero, any thoughts from the community on how to proceed with this/any workarounds on this situation? Thanks in advance.

Link: Keracyanin cation | C27H31O15+ | CID 441674 - PubChem

2

Most parameterization methods work with a net charged molecule, so that shouldn’t be a problem. But in most (almost all) cases you will want your simulation system to be net neutral, but that is easily achieved by adding counter ions.

3

Thanks for the reply, but my problem is not with the net charge, The problem is that one of the oxygens has one extra bond, which is messing up the overall valence of the compound. This is leading to crashing of all ff parameterization programs I’ve tried so far.

4

I only tested the SDF file from PubChem and it works fine with AcPype/Antechamber via STaGE at least (presumably without STaGE as well). I didn’t check the quality of the output, leaving that to the domain experts, but I could generate a GAFF2 topology (with net charge 1).

1 Like