Bill wrote: > Your question regarding surfaces reminded me of an inconsistancy I > reported some time ago.
Yes, I now recall that you pointed this out some time ago. > Chime and Rasmol display a CPK or Spacefill > (100%) image that closely resembles a Pauling (CPK) model. The "dots on" > surface that I sometimes use to surround a stick model is the same as > the CPK surface. In Jmol, however, the CPK or Spacefill display has > hydrogen atoms more deeply imbedded in larger atom spheres. You can see > this in a sample display of methane at: > http://www.cem.msu.edu/~reusch/VirtualText/methanes.htm > <http://www.cem.msu.edu/%7Ereusch/VirtualText/methanes.htm> > OK, I spent some time investigating this ... but I have more questions than answers. First, some background and some facts: Per the document at http://www.umass.edu/microbio/rasmol/rasbonds.htm ... Most PDB files do not contain H atoms. Therefor RasMol (and presumably Chime) defines something called a 'United Atom Radius' for Carbon, Nitrogen, Oxygen, and Sulfur. These radii are larger than the normal vdw radii because they, at some level, incorporate the hydrogen. RasMol uses either the vdw radius or the united radius, depending upon whether or not hydrogen atoms are present in the PDB file. That is, if there are hydrogens present in the file then RasMol uses vdw Radius. If no hydrogens are present then ituses the larger united atom radius. It is unclear to me what RasMol's behavior is for non-PDB files. Because of this, I would expect the RasMol spheres to be *larger* than the Jmol spheres ... but we observe the opposite. I looked at the values for vdw radii used by Jmol. The table below compares a few elements: Element RasMol vdw RasMol united Jmol/OpenBabel vdw ------- ---------- ------------- ------------------ H 1.100 N/A 1.200 C 1.548 1.872 1.950 N 1.400 1.507 1.850 O 1.348 1.400 1.700 P 1.880 N/A 2.080 S 1.808 1.848 2.000 We can see that the Jmol vdw values are larger than the RasMol vdw and the RasMol united radii for all of these elements. I took the Jmol values from OpenBabel (as documented and referenced in the Jmol source code). I do not know where they came from. However, now that I think about it, it is not clear to me whether or not these values in OpenBabel are even used ... Certainly the covalent radii are used, but the vdw radii may not even be used. It now seems to me that Jmol should take these vdw radii from a standard reference of vdw radii ... like the CRC, perhaps? Q: Is there a 'standard reference' for VDW radii? Perhaps the CRC? Q: Do people agree that we should use these values as the defaults? > I do not have any expertise in this area, but I suspect that Jmol may be > using true vanderWaals surfaces, whereas Chime and Rasmol are trying to > duplicate the Pauling models. Since Pauling models are commonly used by > chemists, I have taken to specifying Spacefill 65% to 70% in most of my > models. Unfortunately, the dots surface cannot be adjusted in the same > manner. Assuming that we get the issue above resolved, and that Jmol starts using some 'published/standard' vdw radius table ... Q: Bill, do you believe that there is a 'Pauling' radius which is substantially different from the VDW radius? Miguel ------------------------------------------------------- This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting Tool for open source databases. Create drag-&-drop reports. Save time by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc. Download a FREE copy at http://www.intelliview.com/go/osdn_nl _______________________________________________ Jmol-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jmol-users
