Ok - so just for my education - the lesson here is that even if you
explicitly define Hs (in a MolFromSmiles manner, C[NH]C ), the hydrogens
are not represented in the underlying graph.
Correct? And this is why I do not see H as a neighbour of N.
Only when we do AddHs are the H actually represented.
-
Jean-Paul Ebejer
Early Stage Researcher
On 14 March 2012 10:46, Greg Landrum <[email protected]> wrote:
> Dear Jean-Paul,
>
> On Wed, Mar 14, 2012 at 11:11 AM, JP <[email protected]> wrote:
> > Indeed - as shown in RDKit 2012.01 by:
> >
> > #!/usr/bin/env python
> >
> > from rdkit import Chem
> >
> > mol = Chem.MolFromSmiles("C[NH]C")
> >
> > for atom in mol.GetAtoms():
> > neighbours = [x.GetSymbol() for x in atom.GetNeighbors()]
> > # where is my H dude?
> > print "%s has neighbours %s" % (atom.GetSymbol(), ',
> '.join(neighbours))
> >
> > Output:
> > C has neighbours N
> > N has neighbours C, C # Here, here!
> > C has neighbours N
> >
>
> I would not expect that one to work: the molecule has no Hs in the
> graph. You have to call AddHs() or, if you read from an SDF or mol2
> file, make sure that you don't remove Hs when the molecule is read in
> (that's the "removeHs" argument to MolFromMolFile or MolFromMol2File).
>
> This morning I managed to convince myself that there was a bug here,
> by screwing up atom indices, now that I've I'm doing the right thing
> I've changed my mind:
> In [31]: mol = Chem.AddHs(Chem.MolFromSmiles("CNC"))
>
> In [32]: mol.GetNumAtoms()
> Out[32]: 10
>
> In [33]: [x.GetSymbol() for x in mol.GetAtomWithIdx(1).GetNeighbors()]
> Out[33]: ['C', 'C', 'H']
>
>
> -greg
>
------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Rdkit-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
