I'm attempting to generate a symmetry file for a system comprised of two parallel beta sheets symmetrically related only by translation: the beta strands within the sheets are symmetric translations, and the sheets themselves are symmetric translations. I additionally want to keep the z-axis offset between the beta sheets constant, if possible, so I can change the x- and y- spacing between the beta sheets. I have had success with this in similar systems using the HELIX mode of the make_symmdef.pl program, but in this case there's no proper 2-fold axis, trying to use HELIX mode generates the wrong symmetric structure. NCS mode complains that the chains are only related by translation and won't do anything, CRYST mode is irrelevant, and PSEUDO mode defines a jump between the centers of masses of the sheets, which aren't lined up so there's a chance the z-axis offset will change. If I'm wrong about the PSEUDO behavior, then it should work fine for what I need. Otherwise, if anyone can give me any suggestions on how to set up these symmetry definitions, I would greatly appreciate it.
The HELIX mode of the script should handle this, if it is just a translation. What is incorrect about the structure generated in helix mode? Can you post the structure?
In helix mode, it generates a single beta sheet. I'm attaching three PDBs:
dry_snqnnf_rechained_reordered.pdb: The crystal structure I'm trying to emulate. It's a type 2 steric zipper (amyloid fibril-related structure). Chains A-E are one sheet, F-J are the other.
snqnnf_helix_symm_incorrect.pdb: When I run in helix mode with "-a A -b B". It generates a single sheet. Probably the wrong way to do this.
snqnnf_helix_symm_incorrect_cross_sheet.pdb: When I run in helix mode with "-a A -b F" (the sheets are staggered, that's why I need helical symmetry, the centers of mass are not in the same plane). It generates a single out of register beta sheet which isn't close to the structure. I believe using this configuration with a type 1 steric zipper (such as Sup35 NNQQNY) works without any issues. Trying to use the "-i" flag makes things even messier, I don't think there is any point symmetry in this system.
I tried changing the jumps and DOFs from the pseudo-symmetric output to emulate the helical symmetry output of NNQQNY. I think I'm making a mistake with a jump group somewhere, because either the structure isn't assembled properly or the rigid body moves aren't correct. That's all I can comment on so far.