I've made some interesting observations, with flat MOL file models that have
the (newly
supported by Jmol) stereo bond codes (1 for up, 6 for down):
(All atoms have z=0)
2 programs that have a good support for the MOLfile standard:
ChemSketch opens the files and depicts them flat with the proper wedge bonds.
Accelrys' Discovery Studio Visualizer 2.0 opens the files and depicts them 3D,
so it has done
an optimization behind the scenes, like Jmol does with the
load myFile.mol filter "2D"
method. So this is not unique to Jmol!
However, in the dimethylcyclopentane I just used for testing, DSV retains the
proper stereo,
and does not add hydrogens, while Jmol adds H but looses the stereo config when
minimizing.
Now an idea:
The MOL format spec says that flat models should have a 2D tag*. Up to now,
Jmol is not
parsing it.
My suggestion: check if the 2D tag is present and, in that case, apply the 3D
conversion even
if a plain "load" command is used. That would match in the 2D-MOL files the
behaviour with
JME files.
How does it sound?
*) line 2, columns 21-22 should be either "2D" or "3D"
------------------------------------------------------------------------------
_______________________________________________
Jmol-developers mailing list
Jmol-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-developers
