Re: [OpenBabel-scripting] pybel: Addition of hydrogens

[Noel O'B. <bao...@gm...>]

2008/4/29 Noel O'Boyle <bao...@gm...>:
> ---------- Forwarded message ----------
>  From: Florian Nigsch <fn...@ca...>
>  Date: 2008/4/29
>  Subject: pybel: Addition of hydrogens
>  To: Noel O'Boyle <bao...@gm...>
>
>
>
> Hi Noel,
>
>   And here we go with the next question:
>
>   1) I read in a molecule (sildenafil), remove the hydrogens, get a
> canonical smiles
>  (CCOc1ccc(cc1c1nc(=O)c2n(C)nc(CCC)c2[nH]1)S(=O)(=O)N1CCN(C)CC1) which
>  I keep to create the same molecule again later.
>   2) I read the molecule back in from the smiles with pybel.readstring()
>   3) I delete the hydrogens with OBMol.DeleteHydrogens()
>   4) I match with pybel.Smarts("[R]~[!R]") - i.e., get anything
>  non-ring that is connected to a ring
>   5) print [m.atoms[x-1].OBAtom.GetType() for (y,x) in SmartsMatch] and get
>
>   ['O3', 'So2', 'O2', 'C3', 'C3', 'So2', 'C3'] - which is fine.
>
>   If I leave out step 3, then I find
>
>   ['O3', 'So2', 'O2', 'C3', 'C3', 'H', 'So2', 'C3']
>
>   and there is suddenly a hydrogen in there. Where does it come from?

I note that the canonical smiles in step 1 also contains a hydrogen. I
have a feeling that the hydrogen is necessary to specify the
protonation state of the N in the ring. Maybe someone else can
comment.

>  Does OpenBabel add hydrogens internally when reading in from a smiles?
>  Why does it only match one out of all possible ring-hydrogens (there
>  are about 8 I think)?
>
>   Thanks a lot,
>
>   Flo
>