Hi Bob, Thanks very much for all your suggestions! I particularly like the "isosurface cavity" command, which does exactly what we were trying to do. It confirms again, that the "protein people" are one or two steps ahead of us normal mortals...
I tried it, and it is almost perfect. There are of course small differences between a protein and a zeolite. A zeolites can have cavities (or cages as we call them) that are not open to the outside, i.e. are not part of the channel system. To be perfect for me, I would need the possibility to exclude cavities that are "closed" for a given size of an atom. In order to make it clear what I mean, look at the example on: http://crystals.ethz.ch/IZA-SC/3D-images/JmolViewer_ch.php? cifname=EMT&FTC=EMT&acell=%2017.215&bcell=%2017.215&ccell=% 2028.082&alpha=%2090.000&beta=%2090.000&gamma=%20120.000 and select "cavities" under "Surface Display" When you turn the model round, you will see small cages (cubo- octahedra) that have an unconnected green ball in the center of a cage. I do not know, how easy it is to check for a "cavity" whether it can be entered by a sphere of a given radius (e.g. 2A) or not. If this is possible and you can open the channels, as you offered, that would make me very happy... If this is not possible, I can work as you suggested with the command "select within...." . I checked this too and it seems to me I can do a nice box by making a series of selections with "within( distance, HKL plane)" . So, I will still be happy. Thanks again for your help. Christian P.S. A way round the "closed cavity problem" would be, that I fill them with some dummy atoms. I guess, I could do that. On Apr 14, 2007, at 5:59 AM, Bob Hanson wrote: > Christian, > > There really should be no need for two models. If you have the > capability to build the models with an overabundance of atoms, that's > even better. Just use that larger model. use clipping planes or even > just > > select within (0.5, cell=555) > > to fine tune the atoms you want, then you will see the open tubes. > It's only if you include > > isosurface..... ignore (xxxx) > > that you will see the closed ends then. > > > On 4/13/07, Christian Baerlocher <[EMAIL PROTECTED]> wrote: >> Hi Bob, >> >> Thanks for your help. I think I understand now how it works. It is >> indeed tricky, but I think I can get it to work. >> >> However, I still have a few problems: >> >> I cannot use the cell selection, as this is too crude. I need to >> specify fractions of cells, i.e. I may need boundaries such as {-0.2 >> 1.2 -0.3 1.3 0 1.0} in order to get nice tubes. So I have >> written a small program that gives me two models, one with all atoms >> within the desired range and one with only atoms outside this >> boundary. By selecting only the "inside" atoms of the first model I >> get the desired open tubes. This works, but maybe there is a cleverer >> way to do this. The openings are a bit jagged, but I think this >> cannot be changed. > > the jaggedness will be there, but you can make it a bit smoother by > changing the isosurface resolution. > >> >> So far so good. Now the problem: in order to get the channel system, >> we fill it with spheres that have different radii. I want now to >> calculate the surface around these "atoms", so I need to specify >> their radii in Jmol. I tried it with spacefill, which works fine for >> the atom display, but these are not used in the calculation of the >> isosurface. I need to define the VDW radii myself, but I did not find >> a way to do this. Is there one? > > No, but that would be a fine use of the DATA statement. > > Consider using > > isosurface cavity > > It won't be open, but it will be the exact surface around your > "atoms". This algorithm inserts the exactly fitting sphere for a > cavity at each 3D grid point. The result is a set of overlapping > spheres of the radii you describe. Then the surface is drawn around > those spheres. Should be perfect. I think I could write something that > would open it up on the outsides if that is the only thing you need to > make isosurface cavity do what you need. > > By the way, the voxel count in that cavity gives you a measure of its > volume pretty closely, I suspect. > > > >> >> Of course, I will be glad to send you the result of my efforts. So >> far I have only an experimental page which is not yet properly >> connected to the zeolite database and which shows no channel surface >> yet. :-) >> look at >> http://crystals.ethz.ch/IZA-SC/3D-images/JmolViewer.php? >> cifname=LTA&FTC=LTA&acell=%2011.919&bcell=%2011.919&ccell=% >> 2011.919&alpha=%2090.000&beta=%2090.000&gamma=%2090.000 >> >> to see just one example of a zeolite (all on one line) >> > > lovely. > > ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Jmol-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jmol-users
