I'm trying to create a PSF for a small part of the human vault protein, and I am running into issues doing so.

• VMD seems to not be recognizing the residue names in the pdb. I noticed that the residues had the suffix A (e.g., ALAA vs ALA); I removed that suffix directly in the pdb file, but that did not solve the problem. I also tried using PyMol to process the needed changes, and that did not solve the problem, either.
• When I run the script to generate the psf, VMD says there are duplicate keys for the residues. Columns 23-26, which contain the residue ID in the pdb file, do seem to contain distinct ID numbers for each residue. Since there are multiple atoms in each residue, I would assume that multiple atoms will have the same residue ID.
• The script I am using is supposed to ensure that each chain in the original file is preserved in the psf. However, when I load the finished psf and its corresponding pdb into VMD, it does not look like each chain was preserved. I can tell by opening the "graphical representations" window, going to the "selections" tab, and selecting the "chains" option.
• Additionally, when I load the finished psf and its corresponding pdb into VMD, it appears that only a fragment of the original 8 chains was loaded into the psf. My assumption is that for some reason VMD could not generate the coordinates for the other atoms in the structure, and so left them out.

I have attached the original pdb (frag8_5.pdb), the script (create_psf2.tcl), the topography file (top_all36_prot.rtf), and the output psf and pdb files (structure_per_chain.*).

Any help troubleshooting these issues would be appreciated.