On 9/25/07, Andreas Schüller <[EMAIL PROTECTED]> wrote: > I would definitely expect the SmilesParser to return full-fletched, > fully connected molecules.
Minor note: a SMILES does not require one single connected molecule. > Actually, I don't find it intuitive to need a post-processor for the > SmilesParser. > > So the SMILES parser should: > 1. Parse the SMILES into a connectivity table > 2. Resolve the connectivity table and deduce bond orders I did not clearly state that. All bond orders given in the SMILES are, of course, taken for granted and used. The connectivity is not resolved, just taken from the SMILES string. What I wanted to point at, is those situations when SMILES are presented to the algorithm where *no* explicit bond orders are given. > 3. Return a fully connected molecules with correct bond orders The correct bond orders is the problem (if not explicitely given in the SMILES). Egon -- ---- http://chem-bla-ics.blogspot.com/ ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Cdk-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/cdk-user
