Hi all, On Sun, Jul 31, 2011 at 2:12 AM, Robert Hanson <[email protected]> wrote: > On Sat, Jul 30, 2011 at 3:42 PM, Richard Ball <[email protected]> wrote: > Egon, what do you recommend?
R,S stereochemistry is a very complex algorithm. I have a limited implementation in the CDK of the published algorithm, but still we get a good amount of fails. The algorithm is not written by an information specialist, leading to nasty rules that are difficult to implement efficiently. For example, rules that resolve R,S based on the R,S value of other stereocenters, or rules that are based on E,Z stereochemistry. Even more interesting are the rules that state that R,S is decided by the R,S of the atom itself. Like "If I am R, then I am R indeed", rather than "If I am S, then I should be R, which is not possible, so I am not S" :) In short: the rules are nasty. The CDK gets 98.5% right for molecules with one stereocenter, but this goes down for more complex molecules. The lack of a good, Open gold standards is not helping here. That said, other Open Source software that implements (part of) the rules include OpenBabel and OPSIN. With the CDK and the online resolver, we have a few tools to compare against each other. I would advice against trying to implement these rules, particularly because they are mostly needed for coming up with IUPAC names, which Jmol does not do anyway. But then again, I have said that before, only encouraging to you pick up the challenge ;) Grtz, Egon -- Dr E.L. Willighagen Postdoctoral Researcher Institutet för miljömedicin Karolinska Institutet (http://ki.se/imm) Homepage: http://egonw.github.com/ LinkedIn: http://se.linkedin.com/in/egonw Blog: http://chem-bla-ics.blogspot.com/ PubList: http://www.citeulike.org/user/egonw/tag/papers ------------------------------------------------------------------------------ Got Input? Slashdot Needs You. Take our quick survey online. Come on, we don't ask for help often. Plus, you'll get a chance to win $100 to spend on ThinkGeek. http://p.sf.net/sfu/slashdot-survey _______________________________________________ Jmol-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jmol-users
