> Just discovered, downloaded and spent a few hours playing with > Jmol-8/JmolApplet-8 and the �sneak preview� of Jmol-10. ABSOLUTELY > BRILLIANT! Glad that you like Jmol
> The materials I have been working on have been mainly targeted at > pre-university level (AS and A2 level in UK edu-speak) and involve > inorganic crystal and metal structures as well as organic molecules > and biopolymers. All currently use .pdb format. Phillip Barak of the University of Wisconsin USA recently sent me a link that shows some excellent web pages with inorganic molecules. You should take a look at them: http://www.soils.wisc.edu/virtual_museum/ > In this context, I have the > following wish list for RasMol/Chime scripting and other options in > the forthcoming Jmol/JmolApplet releases. > > > 1. set unitcell on/off > > It would be good to be able to display the unit cell when > crystallographic data is present in .pdb files. As set axes and set > boundbox are already implemented I presume this would not be too > difficult. In my experience, the rendering of the unit cell in > RasMol/Chime was imperfect, particularly when it intersected atom > spheres at the corners of a unit cell, so it would be nice if this was > done �properly� in Jmol. This is a bug ... good catch. The Jmol application currently supports unit cell display. But the command was not implemented in the script interpreter. Several of the other Jmol developers are crystalographers, so I suspect more of the 'crystal' functionality will make it into the scripting language over time. > 2. set connect false/save (and initscript parameter) > > For complex structures RasMol and Chime usually ignore CONECT records > in .pdb files and use their own internal algorithms to determine > bonds. This means that structures with �unusual� bond lengths, e.g. > ionic and metal lattices, may not display correctly even if the bonds > are > described by CONECT records. Further explanation and details can be > obtained from: > > http://www.umass.edu/microbio/rasmol/rasbonds.htm > > The workaround for this in RasMol 2.6 (not sure about RasMol 2.7) and > Chime 2 is provided by the set connect false command which forces use > of the CONECT records. In addition Chime 2 has the set connect save > command which forces the plugin to show bonds for all CONECT records > in the .pdb file, while at the same time calculating bonds not > specified in the CONECT records in the usual manner. Set connect save > can also be issued in the embed tag which invokes Chime 2 using the > initscript parameter. Initscript is required to execute the command > before the molecule is loaded. > > It would be nice if the JmolApplet would implement set connect > false/save with initscript, or some equivalent mechanism, to achieve > the display of structures with �unusual� and heterogeneous bond > lengths. Not sure what you mean by 'initscript' ... I have no practical experience using either Chime or RasMol :-) Also, I was unable to find any reference in the RasMol doc to 'set connect' Is this a Chime extension? Or is it just undocumented? Jmol respects atom connectiving in .mol files, but I think it currently ignores CONECT records in .pdb files ... this needs to be implemented. The RasMol doc talks about using two different algorithms for small vs. large molecules. Jmol doesn't do this. We have one algorithm that (we think) is pretty fast. The rules for bonding are (generally) taken from OpenBabel ... and I *think* it does a better job with metals than RasMol ... the iron in hemoglobin hemes is correctly bonded in Jmol but is not bonded at all in RasMol. Specific examples of molecules that are not bonding properly would be helpful. With that said, clearly we need to implement the 'set connect' behavior you are describing so that we respect CONECT records when they are present > 3. scaled > > Chime includes the scale3d embed tag. When the same scale3d value is > used for multiple structure windows on a page, the viewer can compare > the relative size of the structures. I find this very useful for > visualising, for example, trends in atomic/ionic/metallic radii and > bond lengths. Good. > 4. Alternative units (pm and/or nm) for displaying distances (in > addition to �) > > In common with most molecular display programs, Jmol, RasMol & Chime > use � as the unit of distance measurement. Although the �ngstr�m is > the �natural� unit for bond lengths, SI units are recommended for > pre-university students in the UK. Hence these students tend to > encounter bond lengths measured in pm (or nm) in most textbooks, exam > papers etc. I certainly don�t want to start a long thread discussing > the relative merits of different units or at what point in their > education students should be introduced to the �. However, from my > point of view, it would be good to have some mechanism for switching > to different units (pm) for distance measurements, with perhaps � as > the default. > > Of course, I am also looking forward to all the protein rendering > options and other goodies promised for Jmol-10. Again, thanks for all > your efforts. Seems like an OK idea to me. My only reservation is that we need to find a way to add this control without adding too many changes to the scripting language. To date I have been reluctant to implement commands outside the RasMol/Chime set. In the future we will clearly have to do so, but I want to get some end-user input before we start adding additional scripting commands. Thank you *very* much for taking the time to write this. We find this type of list quite helpful. All of these items have been added to our bug database on sourceforge. Miguel > Geoff Rowland > > [EMAIL PROTECTED] ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Jmol-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jmol-users
