Jan answers the question perfectly below (thanks Jan!)

It's certainly possible to modify the RDKit's default valence table as Jan
describes. It requires a bit of testing to make sure that things like 3D
coordinate generation still work.

Given that some of us are not particularly fond of general support for
hypervalent species, it is probably worth thinking about adding a method to
allow the default valences for an atom to be modified at runtime. This
would solve this problem (well, not the 3D coordinates problem) in a
general way and allow arbitrary flexibility.

-greg


On Friday, 29 January 2016, Jan Holst Jensen <[email protected]> wrote:

> On 2016-01-29 17:02, Rich Lewis wrote:
>
> Dear all,
>
> I tried to load a periodate from its smiles (I think these are correct),
> but it failed as according to rdkit, Iodine cannot have a valence of 7.
> However, it can when stabilised with oxygens (at least from what I recall
> from my undergraduate course!).
>
> In [*2**3*]: Chem.MolFromSmiles('[O-]I(=O)(=O)=O')
> [15:29:04] Explicit valence for atom # 1 I, 7, is greater than permitted
>
> In [*24*]: Chem.MolFromSmiles('OI(=O)(=O)=O')
> [15:24:21] Explicit valence for atom # 1 I, 7, is greater than permitted
>
> In [*25*]: Chem.MolFromSmiles('[H]OI(=O)(=O)=O')
> [15:24:26] Explicit valence for atom # 1 I, 7, is greater than permitted
>
> I don’t know if the file rdkit/Code/GraphMol/atomic_data.cpp is where the
> valences are configured for rdkit, but in the file, I saw that the allowed
> valence states are +1, +2(?) and +5 but not +7.  A few of the other
> elements are also similarly conservative, such as Br only allowing +1.
> Should these be expanded, or is this deliberate?
>
>
> Hi Rich,
>
> The file rdkit/Code/GraphMol/atomic_data.cpp is indeed where the valences
> are configured. There was a similar discussion in 2015 that resulted in
> additional valence data being added to P.
>
> http://sourceforge.net/p/rdkit/mailman/message/34131420/
>
> You can try out the valence change locally by adding it to atomic_data.cpp
> (TAB-delimited format) and rebuilding RDKit.
>
> Cheers
> -- Jan Holst Jensen
>
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
_______________________________________________
Rdkit-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Reply via email to