Re: [Jmol-users] rendering artifacts in 1hbk

[Miguel <mi...@jm...>]

Jan wrote:
> Hi Miguel,
> I have taken a closer look at 1hbk, generating the -x,-y,1/2+z
> transformation,
> http://www.imb-jena.de/cgi-bin/SCOPlnk.exe?FILE=1hbkZ.pdb&JMOL=1hbk&USER=jcb1.imb-jena.de
> there the HOH46 atomno=865 gets near the Ni1091 atomno=819 of the
> asymmetric unit
> this violates the PDB format guide
> http://www.rcsb.org/pdb/docs/format/pdbguide2.2/part_69.html
> CONECTs to atoms whose coordinates are not in the entry (e.g.,
> symmetry-generated) are not given.

Jan,

OK, so that is why the bonds run all the way across to the other side of
the molecule.

My complaint with the file different.

They are using the hydrogen-bond fields of the CONECT record incorrectly.

> I'll write the RCSB,

Please copy me on their response.

> but my opinion is, that the guide and has to be
> changed not the entry

Q: Which aspect of the guide would you want to change?

The guide has not changed since 1996. And most .pdb files do not comply
with this 1996 standard. It seems that the chance of changing these things
is very small.


> and Jmol should reject unrealistic bonds e.g. >4.2A

This is not going to happen.

Jmol is not a calculations program, it is a rendering program.
Different people have different uses for it.
Jmol will do its best to display *exactly* what is in the files.


Miguel