I've never found a clean way to do this - what I do is to save a PDB file with the original model and the relevant symmetry mates, which I then use for the alignments.  There may be a way to do is with object groups, however.  (FYI, you'll probably also need to rename the chain IDs for the symmetry mates so you can still use a unique atom selection for the original chain(s).)


On Sunday, September 11, 2011, shukunluo <shukunluo@gmail.com> wrote:
> Dear Pymol users,
> I want to compare the crystal packing and contacts of two different space groups of the same protein using pymol. Here's what I did: (1)loaded two structures; (2) aligned two structures; (3) generated symmetry-related objects of two structures. However, the symmetry-related molecules didn't move with the source structure. I tried in coot to superimposed two structures then save the moved molecules, but the space group changed to the target structures. I also tried the matrix_copy command with failure.
> Does anyone know how to do this? Thanks in advance.
> Best,
> Shukun