> We currently have no code to generate hash/wedge flags for a > structure. This isn't difficult for most cases though. I'm doing some > stereochemistry at the moment and could add this to my todo list. (I > haven't looked at the mcdl code in detail but there is some code for > this IIRC) > A question about MCDL and coordinate generation in general - I remember there was discussion to adopt RDKit methods for that, has this been abandoned in favor of MCDL? Not to dis MCDL but RDKit seems to be more flexible (for example it can generate coordinates for any part of the structure while MCDL can only do that for a terminal superatom or the whole structure) and the results appear to be more visually appealing.
> Stereo bonds (i.e. hash, wedge) between stereocenters should be > avoided at all cost. Next, stereo bonds in rings should be avoided > (not always possible). This leaves candidate bonds for each center > that could be used. From here on, coordinates should be used to > generate the "best" looking diagram. > What are you plans regarding user-supplied stereochemistry (hash/wedge bonds)? I hope OB will not reject or modify user-supplied information just because it happens to put a stereo bond in a ring? Igor ------------------------------------------------------------------------------ This SF.net email is sponsored by Make an app they can't live without Enter the BlackBerry Developer Challenge http://p.sf.net/sfu/RIM-dev2dev _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
