How to get a PDB identical to the GIF of the models showed in the I-tasser result tab

Hello!
Some users before me encountered this problem of missing Beta sheets in PDB files from I-Tasser.
What I understood is that I-tasser generates PDBs without secondary structure information. The GIF in the result tab is the result of the JMol "calculate structure rama" algorithm on the PDB without secondary structure ( https://zhanglab.ccmb.med.umich.edu/bbs/?q=node/4096 ) .
PyMol and Chimera (as many others) have default algorithm to detect secondary structure (dssp , ksdssp ) , this is why the PDBs look different in JMol and PyMol/Chimera.
Anyway, I followed the instructions on the previous link to have an exact copy of the GIF showed by I-Tasser in Jmol. They work!
My big problem now is that I can't generate a PDB from Jmol that contains the secondary structures information defined by "calculate structure rama" script.
I tried many of these commands ( http://wiki.jmol.org/index.php/File_formats/Export ), but they don't work, I can just save the initial PDB without secondary structures information. (I can save these secondary structure changes only in Jmol current status)
So, do you know how can I write this pdb containing the secondary structure information derived by "calculate structure rama"?
I really hope that you will answer my question, I am sure that I could be useful for many other users too!
Cheers,
Matteo