Dr. Bernstein, > The internal data structures support your idea, but there are other > issues to resolve. > > Usually the positioning changes done to a molecule in RasMol are > exported in a script, rather than in a coordinate file, because > the program centers molecules on load, so on reloading each > molecule needs to be shifted again. One way around this would > be to add the option of writing out the ensemble with molecules > wrapped in model - endmodel brackets, which would allow the > position changes to be preserved. Another solution would be > to add a mode of molecule loading, or a post loading command > that disables the initial recentering, as well as your suggested > revision to the write command, so that molecular repositioning could > be preserved in the coordinate files and not just in scripts. The > information is there. We just need to settle on the user externals. > > Comments? Would it be possible to have a toggle such as "autocenter <on/off>", and have it so that it works whether the user loads up a protein before or after having issued the command? The command being given before loading a protein is pretty self-explanatory. However, the command being issued after having loaded a protein could reset the same PDB and stick to the file given coordinates. On the other hand, if the user has already modified the standard load view on the software, one other possibility is that of saving a script file of the commands only (none of the headers and rotation type information) to memory, reseting the protein, and reapplying the script. This would once again get the user back to the last view before accepting the "autocenter off" command. At the same time, it would not generate an extra script file that could potentially confuse the user. Have a great day, Gunnar Vikberg