Hi Steffen,

I'm not aware of any way to do this. In the case aromaticity, the generator/parser should preserve this (but it probably doesn't). If it's only the aromaticity you need it might be possible to patch this but if you need other patterns then it is difficult.

Sorry,
J

On 12 Mar 2013, at 23:32, Steffen Neumann <sneumann@ipb-halle.de> wrote:

Hi,

our http://msbi.ipb-halle.de/MetFrag/ uses the CDK
to fragment a molecule (by disconnecting the bonds
in different places). We can then check whether the
mass of the two fragments match what was measured
in tandem mass spectrum. If we now want two write
the fragment structure to disc, we could do that
as e.g. a SMILES.

The problem is that the resulting SMILES might not anymore
be considered a substructure of the full molecule:
If we have an aromatic ring, then a fragment of that
will be written as non-aromatic SMILES.

Question: is there a way to directly create a SMARTS
from a subset of atoms in an AtomContainer ?
The SMARTS could then capture that I am looking
for e.g. three connected aromatic carbons.

Yours,
Steffen






--
IPB Halle                    AG Massenspektrometrie & Bioinformatik
Dr. Steffen Neumann          http://www.IPB-Halle.DE
Weinberg 3                   http://msbi.bic-gh.de
06120 Halle                  Tel. +49 (0) 345 5582 - 1470
                                 +49 (0) 345 5582 - 0
sneumann(at)IPB-Halle.DE     Fax. +49 (0) 345 5582 - 1409


------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
_______________________________________________
Cdk-devel mailing list
Cdk-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/cdk-devel