Re: [Cdk-devel] atom configuration based on atom type (was: UniversalIsomorphismTester differences between CDK 1.0.3 and 1.4.4)

[Nina J. <jel...@gm...>]

On 4 October 2011 16:29, Egon Willighagen <ego...@gm...>wrote:

> On Tue, Oct 4, 2011 at 2:36 PM, Egon Willighagen
> <ego...@gm...> wrote:
> > Ah, a recent patch (well, about a year ago:
> > ae21ee2e39c61ae6d1c6aaf7f8919c0fb40170b3) caused atom typing to be not
> > perceived when aromaticity is not perceived. I puzzled why I was happy
> > with that change... I'll cook up a patch to ensure that atom types are
> > again perceived, even when aromaticity is taking literally from the
> > SMILES string.
>
> OK, this patch actually required the atom typing to be inside the if
> statement, because the atom type perception and configuration caused
> earlier aromaticity information to be lost, caused by line 92 in the
> AtomTypeManipulator.configure() method:
>
> atom.setFlag(CDKConstants.ISAROMATIC,
> atomType.getFlag(CDKConstants.ISAROMATIC));
>
> I have not made up my mind if this instruction is important... what
> d'you think? CDK atom types at least do not associate aromaticity with
> atom types... however, Sybyl atom types do. So, removing this call
> would cause this method to fail for such atom type schemes...
>
>
Do you mean it should not set the atom aromaticity, or not use the flag from
the atom type for this?

Nina


> Egon
>
> --
> Dr E.L. Willighagen
> Postdoctoral Researcher
> Institutet för miljömedicin
> Karolinska Institutet (http://ki.se/imm)
> Homepage: http://egonw.github.com/
> LinkedIn: http://se.linkedin.com/in/egonw
> Blog: http://chem-bla-ics.blogspot.com/
> PubList: http://www.citeulike.org/user/egonw/tag/papers
>
>
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2dcopy1
> _______________________________________________
> Cdk-devel mailing list
> Cdk...@li...
> https://lists.sourceforge.net/lists/listinfo/cdk-devel
>