[Jmol-users] CML visualization in Jmol

[FyD <fy...@u-...>]

Dear All,

I use Jmol to visualize CML files within HTML pages. These CML files are
generated from Tripos mol2 files. This means that the CML files do counta=
in the
connectivities between the atoms.

If you look at
http://www.u-picardie.fr/labo/lbpd/REDDB/uploadfile/F-56/info.html
The second molecule "Fragment RA5" (for instance) presents a connection p=
roblem
and a bond which should not exist is created.
See the corresponding file @
http://www.u-picardie.fr/labo/lbpd/REDDB/uploadfile/F-56/tripos2.mol2.cml

The P(2) atom is linked to O(3) and H(20) to O(19) (this is normal), but =
for
instance O(3) should _not_ be linked to H(20). I am aware that this struc=
ture
presents an un-normal distance (too short) between those O(3) and H(20) a=
toms.
I can understand that Jmol calculates its own connectivities based on ato=
m
distances in the absence of connectivity information, but I would expect =
Jmol
to follow the connectivities if the list of connections is provided in th=
e CML
file (and not to add new ones).

Am I wrong ?

Thanks, regards, Francois