Thanks for the help regrading model-loop-define.py script. I have one doubt in model selection .. The number of residues that has to be mentioned for restraining should according to the target sequence only know ?? ..

Also, I generated some 5 models while restraining some portion of the sequence as beta strands  and after seeing all the structures I found that its model 3 that is having perfectly restrained beta sheets as compared to others but its DOPE score is  more as compared to other structures . In this type of situation how do we have to select the correct model ?? Here's the list of energies that I got :

>> Summary of successfully produced models:
Filename                          molpdf     DOPE score    GA341 score
target.B99990001.pdb          1552.05176   -25341.12891        1.00000
target.B99990002.pdb          1485.22974   -25387.23438        1.00000
target.B99990003.pdb          1559.28723   -25382.83984        1.00000 (perfectly restrained str.)
target.B99990004.pdb          1483.82568   -25213.86523        1.00000
target.B99990005.pdb          1402.83508   -25525.49023        1.00000

Also, I would like to ask one more thing about restraining structure that while giving the residue range generally last 2 or some times 1 residue/s doesnot retain the restrained structure .. In that case what do i need to do ??

After doing the restrained modeling is it required to do loop refinement .. as I am doing loop refinement also by generating some 50 structures and then finally selecting the one with low DOPE score .. Is this strategy right or wrong ??