Hello Thorsten,
just some comments from my side: First: Yes, it would be very useful to
automatically add the hydrogens or to be able to configure the SDF_to_CDK
node to do so. At least, there should be some kind of an error hint if the
hydrogens are missing.
Meanwhile I have tested about 500 compounds and compared the results with
the outcome from another program:
XlogP:
The results are very similar. There is a nearly perfect correlation for the
largest group of compounds. There are two other groups with an shift of
about +0.5 and -0.5 logP units. And there is only one outlier: reference
data -1.2; KNIME 1.3. This was the only example with an logP < 0 in my
sample file.
TPSA:
There is a good correlation for nearly all compounds but not for that one
with sulfur. I suggest to make it possible to configure the node regarding
the inclusion/exclusion of sulfur (and phosphorus).
Number of H bond donors, acceptors, rotatable bonds:
Though these numbers are apparently simple, there are big differences
between results from different programs due to different definitions. For
instance, one can count each single bond (standard in KNIME?) or only that
ones that are "really" rotatable. There are also discrepancies between the
definitions of H bond donors/acceptors by Lipinski and by other authors.
Preferably, one should be able to configure the Molecular Property node
regarding the used definition. At least, one should be able to find out
which definition is used. I have not found this in the help.
With best regards
Christian
P.S.: It would be great if one could calculate 3D coordinates on the basis
of 2D data in KNIME/CDK.
Thorsten Meinl
<[EMAIL PROTECTED]
ni-konstanz.de> An
[email protected]
27.08.2008 08:44 Kopie
Christian Grunwald
<[EMAIL PROTECTED]>,
Michael Berthold
<[EMAIL PROTECTED]>
Thema
Re: Antwort: Re: [Cdk-user]
Antwort: Re: KNIME/CDK
calculations: problems with logP
and TPSA
Hi all,
> This seems to solve the problem: No, there were no explicit hydrogens. I
> have now incuded the "Hydrogen Adder":
>
> SDF_Reader -> SDF_to_CDK -> Hydrogen_Adder -> XLogP -> XLS_Writer
Does this essentially mean, that the logP calculation always needs
explicit hydrogens (yes, the non-chemist speaking ;-))? If so, then it
would be a good idea to automatically add them in the node.
BTW, we will very soon start rewriting the CDK integration (e.g.
adopting to the latest CDK code, automatic 2D coordinate generation upon
converting to CDK, ...). Are there any suggestions you guys have for the
CDK KNIME nodes?
Regards,
Thorsten
--
Thorsten Meinl room: Z815
Nycomed Chair for Bioinformatics fax: +49 7531 88-5132
and Information Mining phone: +49 7531 88-5016
Box 712, 78457 Konstanz, Germany
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Cdk-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/cdk-user
