[Cdk-devel] Re: FingerPrinter

[Christoph S. <c.s...@un...>]

Tharun Kumar Allu wrote:

> Have you thought of any other way of building fingerprints
> may be like using substructure similarity like MDL keys or
> some thing like that.=20

Yes, initially this seemed to be the way to go, but that was before I=20
knew the daylight algorithm, which is extremely elegant. The problem=20
with substructure keys is that they need more computation time and that=20
they are dependent on the choice of the substructure collection.

> Can you please explain me the science behind the fingerprint
> algorithm technique.

> In what way is the pseudo random numbers generated from the
> hash of paths upto length 6 uniquely distinguishes one
> molecule from other?  And why only upto 6?

Have no time for an in-depth explanation, but someone else had :-)
http://www.daylight.com/dayhtml/doc/theory/theory.finger.html

Cheers,

Chris

--=20
Dr. rer. nat. habil. Christoph Steinbeck (c.s...@un...)
Groupleader Junior Research Group for Applied Bioinformatics
Cologne University BioInformatics Center (http://www.cubic.uni-koeln.de)
Z=FClpicher Str. 47, 50674 Cologne
Tel: +49(0)221-470-7426   Fax: +49 (0) 221-470-7786

What is man but that lofty spirit - that sense of enterprise.
... Kirk, "I, Mudd," stardate 4513.3..