ALOGP has been behaving weirdly for some time now

On Thu, May 31, 2012 at 6:10 AM, Kevin Lawson <[email protected]> wrote:

> Hi
>
> I am getting some strange results from the ALOGP descriptor algorithm.
> LICSS uses the following code to prep molecules for descriptor calculation:
>
>                 atomContainer = sp.parseSmiles(s);
>
> AtomContainerManipulator.percieveAtomTypesAndConfigureAtoms(atomContainer);
>
> CDKHueckelAromaticityDetector.detectAromaticity(atomContainer);
>                 CDKHydrogenAdder ha =
> CDKHydrogenAdder.getInstance(DefaultChemObjectBuilder.getInstance());
>                 ha.addImplicitHydrogens(atomContainer);
>
> AtomContainerManipulator.convertImplicitToExplicitHydrogens(atomContainer);
>
> With this preparation the ALOGP descriptor returns 0.0 for both the Smiles
> c1ccccc1 and for c1ncccc1 (benzene and pyridine)
> I also ran the first 1000 compounds from the Wellcome AntiMalarial dataset
> through the calculation.  As you can see from the attachment,
> the values are usually much lower than the quoted values in the data set
> (~4 log units on average) and are, indeed usually obviously too low.
>
> To check that I wasn't making a mistake with the submission code, I ran
> all 1000 compounds through Rajarshi's CDK Descriptor calculator:
> http://rguha.net/code/java/cdkdesc.html with identical results
> (Rajarshi's calculator also returns 0.0 for benzene).  I have also
> converted the
> compounds to Kekule form using the FixBondOrdersTool.  Results are
> different, but no better! (also in the attached file).  I have tried
> changing the
> order of the preparation steps but it doesn't make any difference to the
> results.
>
> Any thoughts?
>
> Best wishes
>
> Kevin
>
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Cdk-user mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/cdk-user
>
>


-- 
Rajarshi Guha | http://blog.rguha.net
NIH Center for Advancing Translational Science
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Cdk-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/cdk-user

Reply via email to