Multiple domains, disulfides and constraints

Hello

I have searched the forum and found information that is relevant to my question but has not helped me reach any resolution

I have a protein that has 4 subunits, 3 linker regions and SRP; it is 379 AA in total.

Our primary focus is on generating models that show known disulfide bonds within each domain, as well as other potential interdomain disulfides that may form in the full protein

For the individual domains we are trying to generate models with I-TASSER that show disulfide bonds in known locations using constraints, however so far none of these models reflect known disulfide bonding. This is the model without constraints:
http://zhanglab.ccmb.med.umich.edu/I-TASSER/output/S81742/

These two are with constraints
http://zhanglab.ccmb.med.umich.edu/I-TASSER/output/S84518/
http://zhanglab.ccmb.med.umich.edu/I-TASSER/output/S83791/

And this is the full protein
http://zhanglab.ccmb.med.umich.edu/I-TASSER/output/S81996/

We do have PDB files for the 4 individual domains generated from another system. You appear to have many other tools on your site that might help make this into a single model but I am not clear as to which one use. Also, as the models do not have linker regions, how do I incorporate these?

Thanks in advance for any advice

David