|
From: rich a. <che...@ya...> - 2004-11-30 19:08:26
|
I have some newbie-type questions about how CDK perceives stereochemistry. These questions arose during efforts to update the CDK-Octet compatability layer (CDKTools - http://octet.sourceforge.net). (1) I'm uncertain about the meaning of the Bond "stereo" field. I've looked at the javadocs for Molecule, Bond, and CDKConstants - and browsed around other documentation but I found no clarification. For example, when I go: Molecule molecule = ...; // some molecule int someStereoField = 234; ... // building the molecule molecule.addBond(0, 1, 2.0d, someStereoField); ... // finish building how is the someStereoField interpreted? Are clients free to use any integer they choose or are the values limited to a certain range? Is this field intended to represent solely E/Z stereochemistry or are other uses such as wedge/hash allowed? (2) Similarly, I'm confused about AtomParity. In a recent thread: http://sourceforge.net/mailarchive/message.php?msg_id=8651094 I found some information. Let's say I have an AtomPairity defined by: central Atom = atom #0 first Atom = atom #1 second Atom = atom #2 third Atom = atom #3 fourth Atom = atom #4 parity = +1 Then how can I use this information to mentally re-create the configuration around atom #0? The MDL molfile spec shows one way to do this, as does the SMILES spec - are either of these procedures used in CDK? Also, what can be said about the rules clients must follow to specify an AtomParity? If, for example, "third Atom" is null, but "fourth Atom" is not, is this a legal construct? What if "central Atom" is not specified? many thanks, r __________________________________ Do you Yahoo!? Read only the mail you want - Yahoo! Mail SpamGuard. http://promotions.yahoo.com/new_mail |