Hi Christian, Just compiled it and tested on my own test set. The results are consistant and according to the expected results. I presume some people would not expect to have rings in the linkers but from the CDK news vol. 3/2 I understand this is the intended behavor. To avoid questions about it it might be wise to include this in the documention.
Thanks for all your help, Peter -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 17, 2007 12:44 PM To: Peter Maas; [email protected] Subject: Re: [Cdk-user] Murcko Frameworks Hi, I just committed the corrections. At least for the test cases are linkers are found correctly. Please pay attention to the linker smiles. When the ring atoms are aromatic the linker smile could look like eg cCCc means small letters, which makes no sense when you want to create molecules out of it. I included a smilesToUpperCase variable to change cCCc into CCCC. best regards Christian -------- Original-Nachricht -------- Datum: Tue, 17 Jul 2007 10:17:44 +0200 Von: "Peter Maas" <[EMAIL PROTECTED]> An: [email protected] CC: [EMAIL PROTECTED] Betreff: Re: [Cdk-user] Murcko Frameworks > Hi Christian, > > Just checked and you're right. For a zero-atom linker Murcko draws 2 > atoms. > For one-atom linker 3 and so on. So he's including the atom of the > ring as well. > Currently the code is only doing this for the zero-atom linker, which > is why I was confused. > If it would be implemented like (so including the ring-atom) then it > would also solve the inconsistency of the exocyclicDoubleBonds > parameter. I agree on keeping the double bonds attached. By the way, > in part 2 of Murcko's article he's doing the same for the side-chains > but he puts a lable on it (Augmented side chains). > > Kind Regards, > > Peter > > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Sent: Tuesday, July 17, 2007 8:45 AM > To: Peter Maas; [email protected] > Subject: Re: [Cdk-user] Murcko Frameworks > > Hi, > > That is, for zero-atom linkers I found 2 atoms. > Yes, these are the two ring atoms. In the Murcko paper a linker is > defined as the atoms between two ring systems + the ring atoms. I > think I have to check if also for the other linkers the ring atoms are > included...I am not sure. > > When exocyclicDoubleBonds = false it will give you a 1 atom linker back. > > I assumed the exocyclicDoubleBond was intended for keeping double > > bonds on a ring fragment in stead of the linkers. > I will have to change the documentation. I wanted to keep double bonds > on ring systems and linkers, better on the Murcko fragment.In some > newer studies these Murcko-like fragments (chemotypes) shows a greater > usefullness in classifying compounds, see e.g. studies of G.M.Maggiora > (and in-house). > So I implemented this instead of the "simple" Murcko scaffold. > I will put it on the to do list. > > best regards > > Christian > > > -------- Original-Nachricht -------- > Datum: Mon, 16 Jul 2007 17:02:06 +0200 > Von: "Peter Maas" <[EMAIL PROTECTED]> > An: [email protected] > CC: [EMAIL PROTECTED] > Betreff: Re: [Cdk-user] Murcko Frameworks > > > > > Hi Christian, > > > > Thanks for you quick fix. > > Obviously I got the same warnings as Egon got but I am able to > > compile the code within the CDK. It does seem to work as intended like this. > > > > I did find some unexpected results in the linkers. > > That is, for zero-atom linkers I found 2 atoms. > > Another confusing result is: > > when exocyclicDoubleBonds = true > > And the is an one-atom linker with a double bond it will give you a > > 2 atom linker back. > > When exocyclicDoubleBonds = false it will give you a 1 atom linker back. > > I assumed the exocyclicDoubleBond was intended for keeping double > > bonds on a ring fragment in stead of the linkers. > > > > Thanks again, > > > > Peter > > > > -----Original Message----- > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > > Sent: Monday, July 16, 2007 10:21 AM > > To: Peter Maas; [email protected] > > Subject: Re: [Cdk-user] Murcko Frameworks > > > > Hi, > > > > I checked your remarks and corrected the linker related problems. > > I cannot submit the changes, because I don't have an actual svn > checkout. > > I > > attached the revised GenerateFragment class, maybe Egon or someone > > else could commit this please? > > I didn't add Murcko side chains and in the moment I don't plan to do it. > > > > best regards > > Christian > > > > -------- Original-Nachricht -------- > > Datum: Wed, 11 Jul 2007 17:13:30 +0200 > > Von: "Peter Maas" <[EMAIL PROTECTED]> > > An: [email protected] > > Betreff: [Cdk-user] Murcko Frameworks > > > > > Hi All, > > > > > > I'm trying to use the Murcko frameworks which are part of the > > > org.openscience.cdk.tools.GenerateFragments. > > > > > > getMurckoFrameworksAsSmileArray() gives me the expected frameworks. > > > getLinkerFragmentsAsSmileArray() does NOT give me the expected > linkers. > > > Actually it provides some of the frameworks. It does not give all > > > the frameworks provided by getMurckoFrameworksAsSmileArray() but > > > it skips a few. > > > Apparently it ignores the zero-atom linker fragments. > > > > > > Is this intended behavior? > > > Is anybody planning to add the Murcko side chains as well? > > > > > > Thanks, > > > > > > Peter > > > > > > > > > > > > > > > ------------------------------------------------------------------ > > > -- > > > -- > > > --- 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 > > > [email protected] > > > https://lists.sourceforge.net/lists/listinfo/cdk-user > > > > -- > > GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS. > > Alle Infos und kostenlose Anmeldung: > > http://www.gmx.net/de/go/freemail > > > > > > > > -------------------------------------------------------------------- > > -- > > --- 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 > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/cdk-user > > -- > GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS. > Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail > > > > > ---------------------------------------------------------------------- > --- 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 > [email protected] > https://lists.sourceforge.net/lists/listinfo/cdk-user -- Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten Browser-Versionen downloaden: http://www.gmx.net/de/go/browser ------------------------------------------------------------------------- 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 [email protected] https://lists.sourceforge.net/lists/listinfo/cdk-user
