One or more copies of several ($\geq$ 2) models

1. Prepare an ASCII file containing h, k, l, Fo,$\sigma$(Fo). Give it the name data.hkl
2. Get a copy of the each of the PDB files containing the coordinates of your search models. Edit the CRYST cards, and give unit cell dimensions and space group number of the target structure. Give them the names model1.pdb, model2.pdb, ...
   NOTE WELL : You will need to have separate files for each and every model present in the asymmetric unit of the target crystal structure, even if these models correspond to the same molecule. To make this clear : suppose that the target crystal structure contains two copies of molecule A and two copies of molecule B. Then you need four PDB files : model1.pdb and model2.pdb are identical and contain the coordinates of molecule A. model3.pdb and model4.pdb are identical and contain the coordinates of molecule B.
3. Do Qs -reso <low> <high> -auto <xxx>, where <low> and <high> are resolution limits, and <xxx> is the total number of models (including multiple copies of the same molecule) that are present in the asymmetric unit of the target structure (for the example above, Qs -reso 20 4 -auto 4). Check that the program goes through the initialisation phase and starts the actual minimisation.
4. Start a proper batch job (and forget about it for a few days ? weeks ?).