On Jun 11, 2007, at 9:20 AM, sushil ronghe wrote: > Therefore a fast bistring comparision to rule out most > structures (e.g. has it got an aromatic ring represented in the > bitstring) before the more deep substructure matching would be very > beneficial. > > This is absolutely fantastic idea. > But what would be the benefit of that kind of search. > If some one really want to use the information of the atoms given > in MOL > file we can use the atomtype class. > > Also if we are using the fingerprinting CDK need to Process the > molecule. > and same for substructure search. SO i don't think there will be > any great difference > i think!
I think the idea is to have the bit strings precalculated for a set of molecules. Then rather than doing a substructure search over all the molecules, you can do a fast bit comparison, and get a smaller number of molecules to perform the substructure search on. That is the way we do it in our databases here - definitely saves time! ------------------------------------------------------------------- Rajarshi Guha <[EMAIL PROTECTED]> GPG Fingerprint: 0CCA 8EE2 2EEB 25E2 AB04 06F7 1BB9 E634 9B87 56EE ------------------------------------------------------------------- There is no truth to the allegation that statisticians are mean. They are just your standard normal deviates. ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Cdk-user mailing list Cdk-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/cdk-user
